Full text of "Earthquakes in Kern County, California, during 1952. A symposium on the stratigraphy, structural geology, and origin of the earthquakes; their geologic effects; seismologic measurements, application of seismology to petroleum exploration; structural damage and design of earthquake-resistant structures"

See other formats

^mimmsi^m^mmmmt'i

»S).va./c.N-.v.y'

STATE OF CALIFORNIA
GOODWIN I. KNIGHT, Governor

DEPARTMENT OF NATURAL RESOURCES

DeWITT NELSON, Director

DIVISION OF MINES

FERRY BUILDING, SAN FRANCISCO 11
OLAF P. JENKINS, Chief

SAN FRANCISCO

BULLETIN 171

NOVEMBER 1955

EARTHQUAKES IN KERN COUNTY
CALIFORNIA DURING 1952

(A symposium on the stratigraphy, structural geology, and origin
of the earthquakes; their geologic effects; seismologic measure-
ments, application of seismology to petroleum exploration; struc-
tural damage and design of earthquake-resistant structures.)

Prepared Under the Direction of

OLAF P. JENKINS

GORDON B. OAKESHOTT, Editor

Price $4.00

CONTRIBUTING AUTHORS

Hugo Benioff
Revoe C. Briggs
J. P. Buwalda
William K. Cloud
G. H. Davis
T. W. Dibblee, Jr.
Beno Gutenberg
H. B. Hemborg
Mason L. Hill
G. W. Housner
Robert L. Johnston
Donald H. Kupfer
Stewart Mitchell
Donald F. Moran
Samuel B. Morris
Siegfried Muessig
Frank Neumann
Gordon B. Oakeshott

G. A. Peers
0. W. Perry

Dorothy H. Radbruch
C. F. Richter
Pierre St. Amand
J. Sch locker
Maurice Sklar
George I. Smith
J. L. Soske
Karl V. Steinbrugge
Harold C. Troxell
V. L. VanderHoof
Archer H. Warne
Robert W. Webb
George N. White
C. A. Whitten
11. D. Wilson, Jr.
G. F. Worts, Jr.

COOPERATING AGENCIES

American Society of Civil Engineers

California Division of Highways

California Division of Water Resources

California Institute of Technology

Department of Water and Power, City of Los Angeles

Intex Oil Company

Pacific Fire Rating Bureau

Pacific Gas and Electric Company

Richfield Oil Corporation

Southern Pacific Company

Stanford University

Union Oil Company

United States Coast and Geodetic Survey

United States Geological Survey

University of California at Santa Barbara

Western Gulf Oil Company

LETTER OF TRANSMITTAL

To The Honorable Goodwin J. Knight
Governor of the State of California

Dear Sir: I have the lionor to transmit herewith Bulletin 171, Earthquakes in Kern County
California during 1953, prepared under the direction of Olaf P. Jenkins, Chief of the Division
of Mines. The Arvin-Tehaehapi earthquake of July 21 and its aftershocks, second most destructive
earthquake in California history, violently disrupted the ground surface and man-made struc-
tures, killed 14 people and did an estimated $60,000,000 damage in the heart of California's sec-
ond greatest mineral-producing county. The volume contains numerous maps showing geologic
features, earthquake intensities, surface ground effects, and structural damage, a large number
of photographs, and papers generously contributed by 36 authorities and 16 cooperating agencies.

Bulletin 171 is not only of great interest because of its thorough account of the most completely
investigated earthquake in our country's history but because the detailed information it contains
will be widely applied in the location and construction of engineering structures such as railroads,
highways, dams, schools, buildings of all kinds, tanks, canals, and power installations, in explora-
tion for petroleum, natural gas, and other mineral resources, and as basic information useful in
establishing building codes and insurance rates.

Respectfully submitted,

DeWitt Nelson, Director
Department of Natural Resources
July 1, 1954

Syciimi.re Canyon, near Arvin, Kern County, just before the aftershock of July 25, 1952. Phoio hy Rolert C. Fiampton.

-f

Dust rising in Sycamore Canyon dui-ing the aftershock of the noon hour, July 25, 1952. Photo hi/ Robert C. Frampton.

CONTENTS

Page
PART I— GEOLOGY 13

1. The Kern County earthquakes in California's geologic history, by

Gordon B. Oakeshott 15

2. Geology of the southeastern margin of the San Joaquin Valley, Cali-

fornia, by T. W. Dibblee, Jr 23

3. Kern Canyon lineament, by Robert W. Webb 35

4. Nature of movements on active faults in southern California, by Mason

L. Hill 37

5. Geological effects of the Arvin-Tehaehapi earthquake, by J. P. Buwalda

and Pierre St. Amand 41

6. Ground fracture patterns in the southern San Joaquin Valley resulting

from the Arvin-Tehaehapi earth<|uake, by Archer H. Warne 57

V 7. Arvin-Tehaehapi earthquake damage along the Southern Pacific Rail-
road near Bealville, California, by Donald H. Kupfer, Siegfried
Muessig, George I. Smith, and George N. White 67

8. Measurements of earth movements in California, by C. A. Whitten 75

9. Effect of Arvin-Tehaehapi earthiiuake on spring and stream flow, by

Revoe C. Briggs and Harold C. Troxell 81

10. Water-level fluctuations in wells, by G. H. Davis, G. F. Worts, Jr., and

H. D. Wilson, Jr ^^

11. Seismic prospecting for petroleum and natural gas in the Great Valley

of California, by J. L. Soske 107

12. Application of seismic methods to petroleum exploration in the San

Joaquin Valley, by Maurice Sklar 119

PART IT— SEISMOLOGY -— 129

1. General introduction to seismology, by H. Benioff and B. Gutenberg— 131

2. The major earthquakes of California : a historical summary, by V. L.

VanderHoof 137

3. Seismic history in the San Joaquin Valley, California, by C. F. Richter 143

4. Seismograph development in California, by H. Benioff 147

5. Seismograph stations in California, by B. Gutenberg 153

6. Epicenter and origin time of the main shock on July 21 and travel

time of major phases, by B. Gutenberg 157

7. The first motion in longitudinal and transverse waves of the main

shock and the direction of slip, by B. Gutenberg 165

8. Magnitude determination for larger Kern County shocks, 1952; effects

of station azimuth and calculation methods, by B. Gutenberg 171

9. Foreshocks and aftershocks, by C. F. Richter 177

10. Mechanism and strain characteristics of the White Wolf fault as indi-

cated by aftershock sequence, by H. Benioff 199

11. Relation of the White Wolf fault, to the regional tectonic pattern, by

H. Benioff 203

12. Strong-motion records of the Kern County earthquakes, by Frank Neu-

mann and William K. Cloud 205

(9)

CONTENTS— Continued

Page

PART III— STRUCTURAL DAMAGE 211

1. Arvin-Tehachapi earthquake — structural damage as related to geology,

by J. Schlocker and Dorothy H. Radbruch 213

2. Earthquake damage to oil fields and to the Palonia cycling plant in the

San Joaquin Valley, by Robert L. Johnston 221

3. Highway damage resulting from the Kern County earthquakes, by

0. W. Perr}-, with supplement. Bridge earthciuake report, Arvin-
Tehachapi earthquake, by Stewart Mitchell 227

4. Damage to water works systems, Arvin-Tehachapi earthquake, by H. B.

Hemborg 235

5. Damage to electrical equipment caused by Arvin-Tehachapi earth-

quake, by G. A. Peers 237

fi. Earthquake damage to railroads in Tehachapi Pass, by Southern Pacific

Company 241

7. Earthquake damage to elevated water tanks, by Karl V. Steinbrugge

and Donald F. Moran " 249

8. Earthquake damage to California crops, by Karl V. Steinbrugge and

Donald F. Moran 257

9. Structural damage to buildings, by Karl V. Steinbrugge and Donald F.

Moran 259

10. The design of structures to resist earthquakes, by G. W. Housner 271

PLATES

Plate 1. Geologic map and sections of southern Sierra Nevada, Tehachapi,

and San Joaquin Valley In pocket

2. Map of the White Wolf fault zone In pocket

References cited in Bulletin 171 279

Finding list of authors 283

Fiiuliu"- list of titles 283

(10)

PREFACE

Tlie major Arviii-Tehachajii cartlKinake of July 21,
1952, and the series of related earthquakes and after-
shocks that followed, are but the latest of a succession
of earthquakes denionstratino- the position of California
in the seismically active belt of geoloffically youngr, de-
velopiufi' mountain ranges and valleys that rims the
Pacific Ocean. The Kern County earthquakes of 1952
accounted for the loss of 14 lives and damape of over
$()(), OOO.OOO in Kern and Los Angeles Counties.

Californians are becoming more earthquake-conscious,
and rightly so, as California and Nevada have had over
90 percent of the earthquakes recorded in the TTnited
States and there is no evidence of any early decline in
earthquake frequency in this area.

It behooves us, then, to be informed on earthquakes :
their origin and geologic causes, their characteristics
and behavior, where they are most likely to occur, their
probable effects in disrupting the land surface and on
engineering structures of all types, their effects on
surface and subsurface water supply, the bearing of
earthquakes and their causative faults on location of
dams, canals, highways, and similar structures, and —
through better understanding — how future losses in life
and property can be reduced. The principal objective
of Bulletin 171 is the presentation of information on
all these things.

Probably no earthquake in history has received as in-
tensive field study by as many scientifically trained peo-
ple as the Arvin-Tehachajn earthquake and the related
aftershocks. Epicenter of the earth(|uake was in the
southern San Joaquin Valle.v, a great petroleum-produc-
ing area, where many geologists are based. Within a few
hours hundreds of geologists from the oil companies, the
United States Geological Survey, California Division
of Mines, Division of Highways, Division of Water Re-
sources, and the universities, were making observations
of surface ground effects along the White Wolf fault,
which was responsible for the earthquake. Just as
quickly, field parties from the Seismological Laboratory
of California Institute of Technology, Pasadena, were in
the area setting up a group of portable seismographs,
augmenting the records obtained at their permanent sta-
tions and obtaining unprecedented coverage of the after-
shocks of a major earthcjuake. Engineering coverage was
also comi^lete, with hundred of engineers and builders
assessing damage to buildings, highways, railroads, and
other engineering structures, and directing repair and
reconstruction.

In effect, the Division of Mines acted as a coordinating
agency for compilation, editing, and publication of this
series of papers dealing with the principal results of
observations and data, from manj- sources, in three main
categories: Geology (Part I), Seismology (Part II), and
Structural Damage (Part III). Field work of the Divi-
sion of Mines consisted in reconnaissance of the geology
of the earthquake area and observations along the White
Wolf fault zone. Through arrangement with the division,
T. W. Dibblee, Jr., mapped the basic geology of over
1,000 square miles of the area as the basis for his paper
on Geology of the Southeastern Margin of the San
Joaquin Valley (Part I, Contribution 2). Drs. J. P.

Buwalda and Pierre St. Amaiid ilid scxeral weeks of
intensive detailed field mapping along the White Wolf
fault zone gathering data for their map and paper on
Geological Effects of the Arvin-Tehachairi Earthquake.
Other papers in Part I comprise discussions of the geo-
logic setting of the earthquakes (Part I, Contribution 1),
fault patterns and characteristics (Part I, Contributions
3, 4, 6, 8), geologic effects along the railroad (Part I,
Contribution 7), effects on water levels and flow (Part

1, Contributions 9, 10), and the uses of seismic methods
in petroleum exploration (Part I, Contributions 11, 12).
Part II presents the results of the extensive seismological
observations, computations and conclusions of the Seis-
mological Ijaboratory, headed by Dr. Beno (iutenberg,
and includes papers by H. Benioff, B. Gutenberg, and
C. F. Richter. Other papers in Part II include a general
introduction to the science of seismology (Part II, Con-
tribution 1), earthquake history (Part 11, Contributions

2, 3) and the results of strong-motion records obtained
by the United States Coast and Geodetic Survey. K. V.
Steinbrugge and Donald F. Moran, both structural engi-
neers with the Pacific Fire Rating Bureau, contributed
the results of their extensive study of building damage
to Part III. That part opens with a paper calling atten-
tion to the relation of structural damage to geology and
closes with a technical paper dealing with the design of
structures to resist earthtpiakes; other papers in Part
III summarize damage to specific types of structure and
installation.

In enlisting contributions from the 16 ditferent agen-
cies, selecting the 36 authors, suggesting the subject mat-
ter for the 34 papers, field checking, compiling, and
editing manuscripts to produce this bulletin, we have
kept in mind the place of the Division of Mines as a
public information bureau on matters directly related to
the mineral resources and basic geology of the State. The
earthquake is a geological phenomenon and the extensive,
disruptive, and complex events that occurred during and
following the Arvin-Tehachapi earthquake are effects of a
geological cause — an abrupt displacement along the White
Wolf fault at a depth of a few miles below the ground sur-
face near Wheeler Ridge at 4:52 PDT on the morning of
July 21, 1952. Mining and petroleum geologists know
faults for their importance in localizing mineral de-
posits, for displacing such deposits after their formation,
and recognize their extreme importance to mineral ex-
ploration. Hence, the series of papers in Part I (Geol-
ogy) deals with faults, fault patterns, and fault history.
Many of the major oil fields in the earthquake-affected
area of Kern County are in structural traps with one or
more faults playing a major role in forming the oil pool.
Recognition of a fault system or pattern, then, may be
of great importance in judging the location and char-
acteristics of faults on the alluvium-covered valley floor
and therefore the possible location of an oil field. Simi-
larly, the principles of the science of seismology and
results of study of the seismograph records discussed
in Part II {Scis)nology) has great economic impor-
tance, particularly as applied to petroleum exploration
(Part I, Contributions 11, 12) on the floor of the valley,
and have been responsible for the discovery of major

(11 )

oil fields. Part III {Strucfural Damage), an account of sary. For all of these economically interested scientists

earthquake damage to buildinofs, oil fields, a refinery, and engineers, and also the many people of our State

highways, bridges, water works, electrical installations. wlio have a great curiosity about earthquakes, this vol-

the railroad, water tanks, and to agriculture, brings to ume has been compiled.

the mind of the engineer the importance of a knowledge Gordon B. Oakeshott

of the nature and behavior of the materials on which his Supervising Mining Geologist

structures are based — the soil and rocks. Those who find Division of Mines

themselves responsible for the design of earthquake- Ferry Building

resistant buildings, for setting up building codes, and for San Francisco

fixing insurance rates find that some knowledge of the July 1, 1954

geologic causes and measurement of earthquakes is neees-

( 1'^ )

PART I— GEOLOGY

INTRODUCTION

PART ONE provides the geologic setting; and back-
ground for understanding of the great succession of
earthquakes that disturbed the southern San Joaquin
Valley area for over a year, beginning with the Arvin-
Tehachapi earthquake of July 21, 1952. It also discusses
those effects of the earthquakes which might be con-
sidered primarily geologic and which required geological
investigation in the field. Maps are provided showing
rock formations, folds, and faults, covering an area of
over 1000 square miles, and the geologic history of the
region is discussed, from the earliest evidences hundreds
of millions of years ago to development of the present
landscape. The earth movements produced striking geo-
logic effects, including ground ruptures in the White
Wolf fault zone, landslides, rock falls, ground fractures

in the Valley floor, and interruption of ground water
and stream flow. These effects are described and shown
in numerous maps, diagrams, and photographs in this
section, and results of quantitative measurements, where
available, are included. The extensive, disrupting, and
complex surface effects of the Kern County earthquakes
are viewed in proper relationship to their geologic
origins as traced through the vast periods of geologic
history. The discussions of seismic prospecting call at-
tention to tlie application of this modern method of
locating subsurface geologic structures, so successfully
combined with geologic interpretation in location of oil
and gas fields on the floor of the Great Valley in the
past 20 years.

(13)

CONTENTS

Page

1. The Kern County earthquakes in California's geologic history, by Gordon B. Oakeshott 15

2. Geology of the southeastern margin of the San Joaquin Valley, California, by T. W. Dibblee, Jr 23

3. Kern Canyon lineament, by Robert W. Webb 35

4. Nature of movements on active faults in southern California, by Mason L. Hill 37

5. Geological effects of the Arvin-Tehaehapi earthquake, by J. P. Buwalda and Pierre St. Amand 41

6. Ground fracture patterns in the southern San Joaquin Valley resulting from the Arvin-Tehachapi earthquake,

by Archer H. Warne 57

7. Arvin-Tehachapi earthquake damage along the Southern Pacific Railroad near Bealville, California, by Donald

H. Kupfer, Siegfried Muessig, George I. Smith, and George N. White 67

8. Measurements of earth movements in California, by C. A. Whitten 75

9. Effect of Arvin-Tehachapi earthquake on spring and stream flow, by Revoe C. Briggs and Harold C. Troxell 81

10. Water-level fluctuations in wells, by G. H. Davis, G. P. Worts, Jr., and H. D. Wilson, Jr. 99

11. Seismic prospecting for petroleum and natural gas in the Great Valley of California, by J. L. Soske 107

12. Application of seismic methods to petroleum exploration in the San Joaquin Valley, by Maurice Sklar 119

PLATES

Plate 1. Geologic map and sections of southern Sierra Nevada, Tehacliapi, and San .loaiioin Valley In pocket

2. Map of the White Wolf fault zone In pocket

(14)

1. THE KERN COUNTY EARTHQUAKES IN CALIFORNIA'S GEOLOGIC HISTORY

By Cordon B. Oakeshott •

The rt'ceiit series of eartluiiiakes in the smitiiei'u San
Joaciuin Valley, initiated by the severe Arvin-Tehaehapi
earth(juake of Jnly 21, 1952 and followed by a succes-
sion of lesser aftershocks, is part of the continuing
evidence of the position of California in a seismically
active belt of geologically yoinig, developing mountain
ranges, valleys, and abrupt continental margins that rim
the Pacific Ocean. No part of the surface of the earth
is free from earthquakes but even the short period of
seismograph records (about 50 years) has been long
enough to show that certain areas on the earth's surface
have many times more earthquakes than others. These
areas of greatest earthquake frequency are the regions
of high, actively building mountain ranges, steep con-
tinental slopes, and deep oceanic belts, one such belt
rimming the entire Pacific Ocean and the other extend-
ing discontinuously from west to east through the West
Indies, Mediterranean Sea, and Himalaya Mountains,
and turning southeastward to join the Pacific belt in
the East Indies. California and Nevada, located in the
great circum-Pacific seismic belt, have had about 95
percent of the earthciuakes in the United States.

FAULTS AND THE CAUSES OF EARTHQUAKES

Causes of Earthquakes. Since earthquakes are vibra-
tions transmitted as waves in the materials of the earth,
they may originate from a variety of causes, including
landslides, explosive volcanic activity, movements of
molten rock at depth, natural and artificial explosions,
and abrupt movements of ma.sses of rock along breaks
in the outer part (crust) of the earth. Seismograph
records show that there are deep-focus earthquakes
originating at depths as great as 400 miles, shocks of
intermediate origin at depths of 27 to 150 miles, and
shallow earthquakes whose foci are at depths of less
than 27 miles. All destructive earthquakes come within
the last group and nearly all have been the result of
sudden movements of blocks of the eartli's crust along
breaks called faults. Rock, which makes up the material
of the earth, is elastic and is known to yield to stresses
by slow creep over extended periods of time. Measure-
ments across the great San Andreas fault in California,
for example, show that the block on the east side of that
fault is moving southward with respect to the west
block at a rate of about 2 inches per j'ear. When the
elastic limit of the rock material is exceeded at any
point or friction along an old fault surface is overcome,
an abrupt movement — similar to the 21-foot horizontal
displacement which took place along the San Andreas
fault to cause the San Francisco earth((uake of 190(i —
may take place. The fundamental causes of accumula-
tion of such stresses are not thoroughly inulerstood but
they are known to occur at the unstable margins of
continental platforms and ocean deeps. The number of
variables involved, the lack of long-recorded data, and
incomplete measurements of all observed phenomena
mean that predictions of specific eartlKpiakes in time
and place are impossible.

• Supervising mining geologist, California Division of Mines.

Faults and Fault Tifiies. Since nearly all destructive
eartluiiiakes result frcmi movenmnts along faults, the
locations and characteristics of faults in an earthquake
belt are of particular interest. This is especially true
of active faults, that is, those that have either a his-
torical record of earthquake foci along their courses or
show evidence of geologically Recent (last few thousand
years) movement. Any fault should be considered active
which has displaced Recent alluvium and whose surface
effects have not been modified to an appreciable extent
by erosion.

Faulting takes place in all kinds of rock and dis-
placement occurring at any one time may be anything
from a fraction of an inch to several feet. Along the
course of a major fault, repeated small displacements,
with resulting eartlKpiakes, may take place at irregular
intervals over long periods of geologic time until the
cumulative relative displacement may amount to many
miles. In such situations the fault becomes a fault zone
of shattered and broken rock that may be more than a
mile in width, often with rock formations of widely
different type, structure, and age brought into contact.
The surface along which movement takes place may
ajiiiroach a plane but is usually highly irregular and
marked by the development of sliekensides, breccia
(broken rock), and gouge (clay-like powdered rock).
The fault surface may dip (incline) at any angle from
the horizontal to vertical, but most faults approach the
vertical. Relative displacement of the opposite blocks
along the fault surface may be horizontal, vertical, or
any combination of these. In some faults the character
of displacement may change along the strike (direction)
and may change in geological time. These character-
istics may all differ along a major fault as it displaces
rocks of varying type and structure.

OUTLINE OF THE GEOLOGIC HISTORY
OF CALIFORNIA

Geologically, the long and complex history of Cali-
fornia properly starts with the origin of the earth some
4 billion years ago and progresses through the geologic
eras and periods, culminating in the development of the
present-day landscape. Each division of geologic time
is represented in some part of the state by rock units
dejiosited or formed in that time. The generalized geo-
logic map herewith shows the broad distribution of rock
types. Intensive study and careful mapping of the rock
formations by hundreds of geologists working in Cali-
fornia, particularly in the ]iast 50 years, has gradually
brouglit some understanding of the salient features of
the state's history. The last period of geologic time — •
the Quaternary — is so recent that features of the present
landscape, in addition to the rock formations, reflect the
events of that period.

Having developed the close relationship between
earthquakes and faulting, it is apparent that the epochs
of most active mountain building, when faulting, as
well as folding and volcanism, are going on most in-
tensively, are the times of greatest earthquake activity.

(15)

Earthquakes in Kern County, 1952

[Bull. 171

FlOTRE 1. Fault types common in Onlifornia:
A, Incipient fault, before movement; /{, Normal
fault; fault surface dips toward downdropped
block ; similar to Kern Kiver fault ; C, Reverse
("thriist" if inclination to liorizontal less tban
4f)°) fault; fault surface dips away from down-
dropped block ; similar to I'leito tbrust fault ;
I). Horizontal, strike-slip fault; left lateral
movement (block opposite *>l)server has ntoved
to left) ; similar to Carlock fault (Snn Andreas
fault is horizontal, but with right lateral move-
ment) ; E, Left lateral reverse fault (combina-
tion of movements of C and 7>; similar to White
Wolf fault.

Consequently, in summarizing California's historj- by
eras and periods, particular attention must be paid to
the epochs of mountain-building (orogenic periods), in-
cluding the last — that of the Quaternary period — in
which we live.

Pre-Camirian Eras. The events of pre-Cambrian
time, covering about three-fourths of the record found
in the rocks, are very little known in California. It is
only in parts of the Basin-Ranges and Mojave Desert
provinces in California that rocks of undoubted pre-
Cambrian age are exposed. There the oldest pre-Cam-
brian rocks (Archean) are gneiss, marble, schist, and
quartzite which are metamorphosed types developed
during great mountain building at the close of Archean
time. The areal extent of that orogenic period is un-
known but it was certainly of world-wide importance.
The overlying rock formations of the later pre-Cambrian
(Algonkian) are little metamorphosed.

Paleozoic Era. Rock formations of possible Paleozoic
age are widely distributed in the mountainous regions
of California, but in only a few localities are they well
dated. Marine Cambrian and Ordovician rocks are best
exposed in southeastern California where they include
sandstone, shale, limestone, and dolomite, indicating
widespread seas. Rocks of Silurian age, also marine
limestone, dolomite and shale, are expo.sed in south-
eastern California, in the northern Sierra Nevada, and
in the Klamath Mountains.

Marine Carboniferous and Permian limestone, shale,
dolomite, and conglomerate, with some interbedded
volcanic rocks, are exposed in southeastern California,
in the Klamath region, and in the western slopes of the
Sierra Nevada. In the ]\Iojave Desert, in the Coast
Ranges, in the Peninsular Ranges, and in the Transverse
Ranges are many remnants of more or less metamor-
phosed rock formations which may be Paleozoic in age ;
their dating as such has not been final.

In the Sierra Nevada it is probable that mountain
building took place at the close of the Paleozoic era, as
rock formations of the Upper Paleozoic Calaveras group
were folded and faulted before the later Mesozoic for-
mations were deposited. Less certain evidence of orog-
eny about this time in the Coast, Transverse, and
Peninsular Ranges has been noted by geologists.

Mesozoic Era. Triassic and Jurassic marine sedi-
mentary rocks, with abundant interbedded volcanic
rocks, are widely distributed in California. The dose of
Jurassic time and, in some parts of the State, early
Cretaceous time, was one of the most important periods
of mountain building recognized. It is known as the
Xevadan orogeny and is best dated in the nortliern
Sierra Nevada. Comparable mountain building in the
Coast Ranges region was probably less extensive, but
there is evidence of widespread mountain building
about that time throughout the Peninsular and Trans-
verse Ranges and in the desert basins and ranges.

The long Cretaceous period was a time of extensive
erosion in the Sierra Nevada, during which removal of
a large cover of the older rock formations, which had
been intruded by the Sierran granite, took place. The
shoreline of the Cretaceous seas lay west of the newly
elevated Sierra Nevada and in the Coast Ranges area

Part I]

Geology

\-. ■. •. ■. ■.'#>. •. ■. ■. ■. •. •. ■. ■. •. ■■•■.".'.^■'•i.'.'i'." *,•'.■■.■.•■ ■■•'v'/.vi"-*3
§j.•.•■.•.^orl"^•.•.■."■.■.•.■^^^^^v//^::}^•^;:v:^•c•■■::■^-^■.;;;^

GEOLOGIC MAP OF CALIFORNIA

SHOWING

PRINCIPAL FAULTS

IN RELATION TO

GEOMORPHIC PROVINCES

AND

GENERALIZED GEOLOGIC UNITS

Geomorpriic provinces from Jenkms.Olof P, 1938, Geomorphic mop of Coliformo,
scale I 2,000,000. Geologic units generalized from Jenkins, Olof P., 1938,
Geologic mop of Californio, scole I 500,000

iCfeloccoos scdimcntofy 'ocks

Figure 2

Earthquakes in Kern County, 1952

Geologic time scale and epochs of mountain building in California.

I Bull. 171

Era

Period

Epoch

Approximate age in
millions of years

Mountain building

Quaternary

Recent

0—0.02

Local uplift and continued active faulting.

fMajor epoch of folding, faulting, and uplift, particularly in the Coast Ranges. Transverse
■j Range*, and northern Peninaular Ranges. Principally elevation, tilting, and faulting in
[ Sierra Nevada, Klamath Mountains. Basin Ranges, and Mojave.

(Crustal disturbances building Rocky Mountains may have extended into southeastern
\ California.

[Great period of mountain building known as the Nevadan orogeny; folding, faulting and
1 uplift to form Sierra Nevada, Klamath Mountains. Transverse Ranges, Peninsular
1 Ranges, and many Desert and Basin Ranges. Milder mountain building in the Coast
[ Ranges. Probable initiation of major fault zones.

Pleistocene

0.02—1.

N
O

K
H
V

Tertiary

Pliocene

Miocene

Oligocene

Eocene

Paleocene

1—9
9—28

28—38
38—58
58—75

Cretaceous

75—130

Jurassic

130—155

Triaasic

155—185

Permian

185—210

Probable extensive mountain building at close of Paleozoic era but location and extent not
accurately known.

Carboniferous

210—265

Devonian

265—320

Silurian

320—360

Ordovician

360—440

Cambrian

440—520

•<

550—2100

World-wide mountain building at close of early pre-Cambrian time; extent in California
area unknown, but evidence of its occurrence is in southeastern California.

Origin of earth.

4000 ±

and over part of the Klamath Mountain province great
quantities of mud, sand, and gravel were deposited. In
southeastern California there is some evidence that the
Tjaramide mountain-building period, during which the
Rocky Mountains received their initial uplift in late
Cretaceous and early Tertiary time, extended into Cali-
fornia.

Tertiary Period. Rock formations of the Tertiary
period in the Coast Ranges, western Great Valley, Trans-
verse Ranges, western margin of the Peninsular Ranges,
and the intervening basins indicate the Tertiary was a
time of intermittent advances of seas from the west with
local elevation and folding of different parts of this
large area at irregular intervals. In general, the Eocene
was the epoch of most widespread seas in the area, Paleo-
cene and Miocene seas were more limited, and the Oligo-
cene and Pliocene epochs were times of restricted seas.
The Klamath Mountains, Sierra Nevada, Peninsular
Ranges, and desert basins and ranges were land areas
elevated at times in some places to considerable relief;
they were the areas which, in general, furnished sedi-
ments to the Tertiary seas which lay to the west. The
Miocene epoch was a time of intensive volcanic activity
accompanying deposition of chert and shale in the
Coast Ranges and western San Joaquin Valley prov-
inces ; volcanism also extended into the Coast Ranges
and Transverse Ranges, in some places taking place below
sea level. The Pliocene was a time of more or less re-
stricted seas, with rapid accumulation of sediments
deposited in localized basins and intermittent volcanic
activity occurring widely throughout the State. The

Ventura and Los Angeles basins contain the thickest
series of marine Pliocene sediments known in the world.
In late Pliocene time the crustal unrest began that
culminated about middle Pleistocene with the great
Coast Ranges-Transverse Ranges orogeny.

Q/uaternary Period. The Quaternary period, the last
million years of geologic time, has been a time of great
mountain building, including extensive folding, fault-
ing and uplift, accompanied by intense local volcanic
activity in certain parts of the State. This mountain
building culminated about mid-Pleistocene time and is
especially well-dated in the Santa Barbara- Ventura
region wliere it has been called the Santa Barbara orog-
eny. While the Coast Ranges, Transverse Ranges and
marginal areas were undergoing great folding, faulting,
and elevation, the Sierra Nevada was being re-elevated
along the series of great fault zones on its east front
and tilted westward with minor folding and faulting
along the western slopes. Volcanoes in the Cascade
Range, Modoc Plateau, and desert basins and ranges
were active in Pleistocene time. The higher ranges, in-
cluding the Sierra Nevada, Cascade and Klamath Moun-
tains, were subjected to periodic glaciation during the
Pleistocene.

Recent time, the last few thousand years since the
general melting and retreat of Pleistocene glaciers, has
been a time of continued active faulting, as shown by
the freqiuMicy of earthquakes, of local uplift demon-
strated by marine terraces along the coast and in some
places by river terraces inland, flooding of the lowest
parts of the coastal areas because of the rise in sea level

Part 11

Geology

on melting of the Pleistoeene frlaeiers, and of the devel-
opment of San Francisco Bay and lesser drowned val-
leys along; the California coast. The development of
California's present landscape has taken place during
this epoch. A necessary and normal accompaniment of
the latter stage of this great orogenic epoch is continued
faulting and frequent earthquakes, with the probability
of gradual decline of this type of activity over the next
many thousands of years.

GEOLOGIC SETTING OF THE KERN COUNTY
EARTHQUAKES*

Fault Pattern in Southern California. The pattern
of known active faults in southern California comprises
right lateral, left lateral, normal, and reverse faults in
large number and great complexity. It has been devel-
oped over long periods of geologic time by great north-
south and northeast-southwest-directed stresses of un-
known origin. Geologic evidence suggests some of the
faidting may have begun as long ago as the Jurassic
period ; modern earthquakes prove that accumulating
strains are still being periodically relieved.

The fault pattern is dominated by the northwest-
trending San Andreas fault, essentially right lateral,
with the east block moving relatively south on the order
of 2 inches a year. Total accumulative movement in this
sense since late Jura.ssie time may be on the order of
300 miles, according to some geologists. The San Gabriel
fault, trending southeast from Frazier Mountain, is of
similar type. Good geologic evidence indicates right
lateral movement on the San Gabriel fault since upper
Miocene time has been po.ssibly as much as 15 to 20 miles,
and since middle Pleistocene time 2 to 2| miles. Other
major north- and northwest-trending faults, including
the Kern Canyon, Sierra Nevada, Nacimiento, and Ingle-
wood, are normal or right-lateral normal and all have
been responsible for many earthquakes in historic times.
The northwest-trending Kern River fault, just east of
Bakersfield, is of comparatively limited extent and may
be a simple normal type.

The Garlock-Big Pine fault zone, otfset about 6 miles
by the San Andreas fault, is the most prominent exam-
ple of the northeast-trending fault system in which left
lateral displacement is characteristic. Displacement along
these faults is also measurable in miles. The White Wolf
fault probably had a major component of movement in
the reverse sense (south block elevated) with some left
lateral displacement. Although recognized by geologists
as having had geologically recent movements, the poten-
tial earthquake threat of these northeast-trending faults
was not fully appreciated until the recent series of Kern
Countv earthquakes initiated bv movement on the White
Wolf fault.

The system of reverse faults comprises a large number
of east-west-trending faults which are shorter and much
less continuous tlian the right lateral and left lateral
systems. A fault in this system usually changes radically
in strike along its course, often varies in dip of fault
surface from less than 45° (thrust fault) to nearly ver-
tical, and can rarely be traced continuously for more
than a few miles. Prominent examples are the Pleito
thrust in the San Emigdio Mountains area south of

• Basic data in this section are talten largely from the papers fol-
lowing in Parts I and II of this bulletin.

Bakersfield, the San Cayetano thrust on the south side
of the eastern Santa Ynez Mountains, the Santa Susana
thrust in the Santa Susana Mountains, and the Sierra
Madre zone of reverse and thrust faults along the south
side of the western San Gabriel Mountains. Although
recent movement on many of these faults is quite evi-
dent, no earthquakes have been definitely traced to dis-
placement on any one of the thrusts.

Eock Formations in the Earthquake Area. The
major Arvin-Tehachapi earthquake of July 21, 1952,
had its epicenter near the eastern end of Wheeler Ridge
and the hundreds of aftershocks, continuing more than
a year later, centered chiefly in the area north of the
Garlock and San Andreas faults, south of the Kern
River and from Maricopa as far east as the longitude
of Tehachapi. This area includes the southern part of
the San Joaquin Valley, the adjacent southern end of
the Sierra Nevada, Tehachapi Mountains, and the east-
trending Wheeler Ridge-San Emigdio Mountains.

The Sierra Nevada, Tehachapi Mountains and the
central part of the San Emigdio Mountains are made up
of a complex of crystalline rocks composed largely of
dark hornblende-biotite quartz diorite (a coarse-grained
rock closely related to granite) of Jurassic or early
Cretaceous age with inclusions of rocks derived from
ancient sedimentary series (Triassie or older) which
have been thoroughly metamorphosed to schist, quartz-
ite, and marble.

The crystalline complex is overlain by a series of
marine and continental sedimentary rocks of the Ter-
tiary and Quaternary periods cropping out along the
foothill areas and underlying the San Joaquin Valley.
This series of stratified sandstone, conglomerate, shale,
and related rocks is thin in the marginal mountain areas
but thickens generally southwestward to an estimated
total of about 28,000 feet just north of Wheeler
Ridge. The Tertiary series dips under San Joaquin
Valley with its contact with the underlying crystalline
rock slo])ing southwest at an average angle of about 6°,
steepening to about 20° immediately northwest of the
White Wolf fault. The valley floor is covered by allu-
vium, including some lake beds and quantities of sand
and gravel deposited by the Kern River and lesser
streams.

The exposed 17-mile portion of the White W^olf fault,
extending from the vicinity of Caliente to a point about
3 miles southeast of Arvin, is entirely within the area
of granitic rock, but the fault extends at least an equal
distance southwest under the valley alluvium to Wheeler
Ridge.

Structure in the Earthquake Area. The southern
Sierra Nevada-Great Valley makes up a more or less
rigid block of crystalline rocks which was intermittently
tilted westward during Tertiary and Quaternary time,
with the eastern part of the block elevated to form the
Sierra Nevada and the western part depressed to form
the Great Valley. The eastern base of the Sierran block
is marked by the Sierra Nevada fault zone; it is paral-
leled about 15 miles to the west by the Kern Canyon
fault zone. Both these great faults appear to be of the
normal type. The area southeast of the White Wolf
fault (left lateral reverse type) was elevated to form

Earthquakes in Kern County, 1952

[Bull. 171

Bear Mountain as a southwesterly extension of the
Sierra Nevada. The northwest-trendino; Kern River fault
is a normal fault locally separating the crystalline block
of the Sierra Nevada from Tertiary-Quaternary sedi-
mentary rocks at the western base of the range. Except
for minor complexities, the sedimentary formations dip
southwestward off the Sierra Nevadan block.

The Tehachapi Range is an elevated block, continuous
with the southern Sierra Nevada, which swings westward
into the San Emigdio Mountains which in turn trend
north of west into the Coast Ranges. The Tehachapi is
followed by the great northeast-trending left lateral
Garlock fault. The southern margin of the San Emigdio
has been cut by repeated horizontal movements along
the San Andreas fault with right lateral displacements
totaling many miles. This has moved the northeastern
block in an easterly direction with respect to the south-
western block. The horizontal shearing movements have
been accompanied by intense compression in a north-
south direction, shattering the basement crystalline-rock
block of the San Emigdio Range, intensely folding the
overlying Tertiary sedimentary rocks at tlieir northern
margin and tlirusting them strongly northward along
the Pleito thrust fault toward the San Joaquin Valley.

Numerous wells drilled for oil in the San Joaquin
Valley have revealed some of the structural features of
the sedimentary formations which are obscured by over-
lying alluvium. They show that the generally southwest-
dipping Tertiary-Quaternary sedimentary strata are
displaced by numerous faults. It is interesting to note
that the fault pattern on the valley floor is similar to
that in the marginal mountains; i.e., northwest-trend-
ing normal, or right lateral, faults and northeast-trend-
ing cross faults (left lateral.').

The pattern of faulting in the earthquake area appears
to be genetically related to joints, shear zones, and planes
of weakness in the ancient cr.vstalline rocks which under-
lie the entire region and crop out in the mountains.

GEOLOGIC EFFECTS OF THE ARVIN-TEHACHAPl
EARTHQUAKE *

White Wolf Fault Zone. According to records of the
Seismological Laboratory of the California Institute of
Technology, the major Arvin-Tehachapi earthquake orig-
inated at a depth of about 10 miles at 4:52 a.m. PDT
on July 21, 1952, at latitude 35° 0(K N, longitude 119°
02' W near Wheeler Ridge. It had a magnitude of 7.7
on the Richter scale, making it one of California's great-
est earthquakes. Rupture, originating at that focus, pro-
gressed N. 50° E. along the White Wolf fault plane, or
surface, at a rate approximating 2 miles per second.
Calculations based on seismological records indicate the
fault dips 60° to 66° toward the southeast and that
the major relative movement along the fault plane was
southeast block (Bear Mountain) up, with a lesser hori-
zontal component of movement toward the northeast.
Thus tile White Wolf fault is a left lateral reverse type
with oblique slip movement.

Existence of the White Wolf fault had been known
by geologists for many years; its general trace was
plotted on a geological map as early as 1906. It had
been recognized as having movement in late geological

• Basic data in tiiis section are taken largely from the papers fol-
lowing in Parts I and II of tliis bulletin.

time but was not considered active in the sense of con-
stituting an earthquake threat. The very steep north-
west face of Bear Mountain, the succession of old land-
slides along that slope, and minor topographic features
made it possible to plot the approximate location of the
fault but its characteristics were unknown. Initial rup-
turing on July 21, originating at depth in solid crystal-
line rock, extended rapidly toward the surface and
reached the surface through altered weathered rock,
soil, and old landslide material to form a series of more
or less discontinuous ground cracks, slides, searplets,
small pressure ridges, ground offsets, and "mole-track"
effects generally along the trace of the White Wolf
fault zone. These features were best developed near the
base of Bear Mountain close to the fault trace from the
base of the Tejon Hills, south of Arvin, to Caliente
Creek. Subsequent aftershocks had little effect on the
features which were developed along the fault trace on
July 21.

The absence of fractures and other surface effects
along the probable southwestern extension of the White
Wolf fault under Valley alluvium between Comanche
Point (Tejon Hills) and Wheeler Ridge suggests rup-
turing was absorbed by the deep alluvium in tliis area
and did not reach the surface. Fractures developed along
the base of the hills at Comanche Point and at the mouth
of Little Sycamore Canyon were near-vertical and formed
northwest-facing searplets (steps) up to a foot high.
Off the fault zone in the mouth of Comanclie Canyon,
numerous small mud volcanoes were developed as a
result of lurching in moist alluvium. A prominent frac-
ture zone follows the edge of the alluvium li miles
northeastward from the mouth of Little Sycamore Can-
yon, with foot-high fractures vertical or dipping steeply
southeast, and left lateral movement up to 8 inches.

The greatest and most continuous zone of fracturing
developed along the fault zone for 5 miles from a point
4 miles due east of Arvin to the White Wolf Ranch.
It is here marked by thrust-fault searplets facing north-
west and the peculiar series of pressure ridges (mole
tracks) up to 3 feet high. These are overthrusts, with
the plane of thrusting dipping southeast at low angles
and with movement toward the northwest over the Val-
ley. In some places a series of parallel mole tracks
replaces the single ridge and northwest-trending tension
cracks developed. In general, the plane of movement
dips 5° to 20° southeastward into Bear Mountain.

The zone of fracturing for 6 miles along the fault
northeastward from the White Wolf Ranch to the South-
ern Pacific Railroad tunnels east of Bealville shows a
different type of rupturing. Fractures in this zone are
not exactly on the fault trace, but are en echelon vertical
ruptures with a more northerly trend than the fault.
There are two or three series of north-trending vertical
fractures 1 mile to 3 miles long in the soil with left
lateral offsets (west block moved south) up to one foot,
but without the formation of searplets.

Between the Bealville Road— U. S. Highway 466 inter-
section and Tehachapi Creek, fractures developed in the
shattered dark granitic rock just north of the apparent
trace of the fault. Trend of these fractures is northeast
and evidence of left lateral movement and compression
is found along the fractures and in the contortion of
rails and shortening of the tunnels. On the hill through

Part 11

Geology

whii'h tunnel 5 passes, there are four parallel fraetnres
treiuliiij;- west of nortii for half a mile and dipping
steeply noi'tli. The uphill side slipped down in each ease
so that an open fissure was fornie<l with a searplet up
to 2 feet hi^h faeing upslojie.

There is no geologie evitlenee tiiat the "White Wolf
fault extends north of Teiiaehapi Creek but a large
north-trending craek 1,000 feet long developed across
the divide between Tehaehapi and Caliente Creeks along
the eontaet between the dark granitic rock and Tertiary
conglomerate; the crack dips steeply west with a 3-foot
west-facing scarp.

Frncturcs and LaitcJsIidhig in the Region. Ground
ruptures, distortions, and fractures were developed very
widely in the region, ajiart from those features closely
associated with the White Wolf fault zone. Such fea-
tures, in general, appear with greater frequency and
prominence near the fault zone. Hundreds of rock falls
and landslides of all sizes took place during the earth-
quake of July 21 and more occurred with each after-
shock. The most damaging landslides were those affecting
the highways. The Ridge Route (U. S. Highway 99) was
blocked at several points between Castaic and Grapevine
and rock falls as far from the fault as the San Gabriel
Mountains partly blocked the Angeles Forest highway
between Pasadena and Vincent. The Caliente Creek road
and the Kern River highway were closed by rock slides
for several weeks. The northwest face of Bear Mountain
and canyon walls, like those of Sycamore Canyon, were
major sites of rock and land slides. Evidence of very
large landslides and the hummocks, depressions, and
shattered rock associated with such slides present before
the late series of earthquakes shows the northwest slopes
of Bear Mountain have been subjected to repeated move-
ments in the past.

Boulders up to 10 feet or more in diameter were dis-
lodged from the slopes of Bear Mountain, Bear Valley,
Cummings Valley, and steep slopes a few miles from
the White Wolf fault zone. Some hopped and skipped
down slope for several hundred feet, gouging the sur-
face and leaving very characteristic boulder trails.

Cracking of highway pavement and barriers, and
slumping of shoulders for miles away from the epicen-
ters were particularly damaging. Cracks affecting pave-
ment, and other stationary structures, had a tendency
to be oriented parallel to or perpendicular to the length
of the structure.

Fractures in AUiirium on the Valley Floor. Frac-
tures and cracks in great variety developed on the floor
of the San Joaquin Valley, principally the result of
lurching in the deep water-saturated alluvium. Many
formed scarps up to a foot or so in height and showed
lateral displacement. However, such fractures were dis-
continuous and many showed no consistent direction or
amount of displacement.

Faint surface lines, oriented northwest and northeast,
had been noticed on aerial photographs by petrt)leum
geologists for years previous to the Kern County earth-
quakes. The Arvin-Tehachapi earthquake produced a
large number of minor cracks in the alluvium, often
marked by swirls and loops at their ends, and oriented
in the same patterns as the older features appearing on
photos. It seems likely that these originated through

recurring small movements on an ancient system of
faults in the basement crystalline rocks which underlie
the alluvium and that settling and adjustments of the
alluvium were reflected in the oriented cracks, lateral
oft'set features, and minor sloughs that appear at the
surface.

Movements Measured by V. SI. Coast and Geodetic
Survey. After the July 21, 1952, earthquake, the U. S.
Coast and Geodetic Survey made repeat surveys of their
triangulation and level schemes. Preliminary results of
these surveys indicate that horizontal movements at the
triangulation stations were small but suggest the Bear
Mountain block moved northeast about 2 feet, while
leveling indicates differences in elevation of 3 to 4 feet
at points aiiproximately 15 miles south of Bakersfield
6 to 8 miles south of Arvin. The area to the south has
been uplifted and the area to the north lowered. The
sharpest break appears about 6 miles south of Arvin.
This is approximately on the White Wolf fault at
Comanche Point.

Effects on Spring and Stream Flow. Studies by the
U. S. Geological Survey show that the flow of many of
the streams and springs in a wide area increased as a
result of the Arvin-Tehachapi earthquake, but it is
doubtful that the effect on the recharge areas or on the
permeabilit.y of aquifers is permanent. The temporary
increase was probably due to disturbance of unconsoli-
dated materials in the discharge areas, resulting in the
clearing of existing outlets and opening of new ones.

One of the most spectacular jncreases was in Caliente
Creek. This increased from completely dry at the town
of Caliente to 25 cubic feet per second within a few
days. As far away as Sespe Creek in Ventura County,
that drainage increased from 17 cubic feet per second
on July 20, 1952, to 37 cubic feet per second on Jul.y 31.
However, 88 percent of the Survey's observation points
in the Santa Ynez Mountains indicated no change in
flow. Streams in Los Angeles Covmty showed little
effect. Radical differences in flow characteristics were
noted, even in short distances.

Water-Level Fluctuations in Wells. Well records of
the U. S. Geological Survey from San Diego County to
Butte County show the earthquake of July 21 reflected
as oscillations of water surfaces. The amplitude of fluc-
tuations ranged from 7.34 feet in a well near the White
Wolf fault about 20 miles northeast of the epicenter to
0.012 feet in a well near Twentynine Palms about 180
miles southeast of the epicenter. Fluctuations did not
vary regularly with distance from the epicenter. Most
of the residual changes in wells in the San Joaquin
\'alley were upward ; elsewhere many were downward.
The records suggest major factors in the fluctuations
were compressibility and elasticity of the aquifers;
factors which, in turn, are closely related to the litho-
logic features of the materials.

UNUSUAL GEOLOGIC ASPECTS OF THE
KERN COUNTY EARTHQUAKES

Probabl.v no earthquake in history has been studied
in the field as intensively or by as many geologists and
seismologists as the Arvin-Tehachapi shock of July 21,
1952, and its aftershocks. Accessibility of the area, the

Earthquakes in Kern County, 1952

[Bull. 171

damage done, the striking land surface effects, the speed
with which the Seismological Laboratory of the Cali-
fornia Institute of Technology focused the attention of
its staff and equipment on the area, and the large num-
ber of professional geologists in the state who visited
the earthquake area have all contributed to the store of
information on the earthquake series and its causes.
Some of the unusual aspects are :

1. The fault responsible for the major earthquake
trends northeast. Nearly all California's great earth-
quakes have been associated with movements on north-
west-trending faults, usually the San Andreas or faults
related to it.

2. The fault is of left lateral reverse type ; the prin-
cipal component of movement was reverse, with lesser
left lateral movement. Most of California's great earth-
quakes have probably been caused by strike slip move-
ments on right lateral faults.

3. Actual surface ruptures along a fault responsible
for an earthquake are rare. The mole-track pressure
ridges, linear ruptures with both vertical and horizontal
offset, up-bowing of ground, and track-shortening in
the railroad tunnels constitute a unique complex series
of surface effects in this fault zone.

4. The White Wolf fault, traceable for about 34 miles,
is one of the shortest known to have been responsible
for a major earthquake in California.

5. Magnitude of the Arvin-Tehachapi shock (7.7)
makes it one of the three greatest in California history.

6. Distribution of the large number of aftershocks
shows not only that readjustments took place along the
White Wolf fault, but also that movements were trig-
gered on numerous other faults in the area. The de-
structive Bakersfield earthquake of August 22, 1952, is
an example of this.

2. GEOLOGY OF THE SOUTHEASTERN MARGIN OF THE SAN JOAQUIN VALLEY

CALIFORNIA

By T. W. Dibblee. Jr.«

ABSTRACT

The southern Sierra Nevada ami Tehaohapi Mountains are made
up of a pre-Cretaceous crystalline complex composed of Juras-
sic (?) Plutonic rocks with hornhlende-liiotite quartz diorite pre-
dominatinR, and linear inclusions of Paleozoic ( '.') schists, (piartzite,
and marhle. The crystalline complex is overlain by a Tertiary-
Quaternary marine and continental sedimentary series cropi)inK
out along the foothill areas and underlying the San Joaquin Valley
where the series thickens soutliwestward to an estimated total
of about 25.000 feet just north of Wheeler Ridge. The Tertiary
series dips under San Joaquin Valley with the crystalline-rock
contact sloping southwest at an average angle of about 0°, stee])-
ening to about 20° immediately northwest of the White Wolf fault.

The White Wolf fault, nearly parallel to the Garlock fault and
about IS miles northwest of it, is a major fault traceable from
Tehachapi Canyon southwest along the base of the steep north-
west slope of Bear Mountain for 17 miles, and probably extends
under San Joaquin Valley toward Wheeler Ridge. The .southeastern
block has been elevated on this fault to a maximum displacement
of at least 10,000 feet as indicated by surface and subsurface data,
with the maximum displacement near the mouth of Sycamore
Canyon.

Surface effects along the White Wolf fault zone produced on
July 21, 10.")2, including overthrusting in the mole-track scarplets
formed, shortening of fences and the railroad tracks crossing the
fault, and dips of the more continuous fault-trace ruptures, strongly
suggest thrusting. Seismographic evidence favors a high-angle
reverse fault at dejjth. <Trt>und cracks and small pressure ridges
oblique to the fault trace, and small ground offsets indicate some
left lateral movement.

The White Wolf fault is essentially a reverse fault, locally a
thrust, elevated in the southeast block, with a small left lateral
component of movement. It is more closely related to the Garlock
and Pleito faults than to faults in the northern part of the area
mapped.

INTRODUCTION

The southeastern mar<>:iii of the San Joaquin Valley
and the adjacent mountain area was the scene of the
violent earthquake of July 21, 1952, which severely dam-
aged the small towns of Arvin and Tehachapi in Kern
County. The cause of this major earthtjuake was found
to be a movement on the White Wolf fault at the base
of the steep northwest slope of Bear Mountain as indi-
cated by ground ruptures formed along the supposed
course of this fault.

The topographic base map which most adequately
covers the area which the White Wolf fault traverses is
the 30-minnte Caliente quadrangle, scale 1 inch = 2
miles, issued by the U. S. Geological Survey in 1914.

The geology of the northeastern quarter of the Cali-
ente quadrangle was taken from previous detailed map-
ping done by the writer in 1950 (Dibblee, 1953). The
geology of the northwestern quarter of the quadrangle
is based on mapping by the writer during several week-
ends in 1950, accompanied several days by A. 11. Warne.
The geology of the southern portion of the quadrangle
and northernmost portion of the adjoining Tejon quad-
rangle is based on published maps and reports by Hoots
(1930), Marks (1938), and Wie.se (1949), although a
week was spent in remapping critical portions of these
areas. Two weeks of the present investigation were spent
in the southeastern quarter of the Caliente quadrangle
in the vicinitv of Bear Mountain and southwest into the

• Consulting geologist. Manuscript submitted for publication June.
1953.

Tejon Hills ; as time was limited, tiic mapping is largely
of reconnaissance nature.

Acknowledgments are due the geological staff of Rich-
field Oil Corporation for access to well logs used to
determine the subsurface structure of the top of the
basement complex buried under the San Joaquin Valley.

STRATIGRAPHY
Basement Complex

The pre-Cretaceous basement complex exposed through-
out the southern Sierra Nevada, Tehachapi and San
Emigidio Mountains, and buried under Tertiary strata
in the San Joaquin Valley, is composed of granitic igne-
ous rocks that form the Sierra Nevada granitic batholith.
They range from granite to gabbro ; quartz diorite pre-
dominates. The metamorphic rocks occur within the
granitic batholith as roof-pendants or linear remnants
of a once tremendous thickness of gneiss, schist, quartz-
ite and limestone. The age of the metamorphic and igne-
ous rocks is not definitely known, although the former
are believed to range from pre-Cambrian to early Meso-
zoic, and the latter are directly traceable into the granitic
rocks of late Jurassic age in the north central Sierra
Nevada where they intrude the Upper Jurassic Mariposa
slate and are unconformably overlain by Cretaceous
sandstones and shales. Brief descriptions of the princi-
pal mapped units of the basement complex follow.

Pelona Schist. The pre-Cambrian (?) Pelona schist,
as mapped by W^iese (1950, pp. 12-13), occurs only be-
tween the two branches of the Garlock fault iii the
Tehachapi Range where about 5,000 feet is exposed. The
formation is highly foliated, with prominent cleavage,
and is composed predominantly of dark greenish-gray
miea-chlorite-albite-quartz schist which was probably
metamorphosed from tuffaceous shale.

Biotiie Gneiss. A large mass of gneiss of unknown
but probable pre-Cambrian age is exposed on the north
flank of the Teliachapi Mountains in the vicinity of El
Paso Canyon. This formation is a complex of well
banded biotite-hornblende-quartz-feldspar gneiss, and
numerous injections of massive quartz diorite.

Pampa Schist. In the Cottonwood Canyon area of
the western slope of the Sierra Nevada are several
lenticular and linear pendants of mica schist within
(juartz diorite. The schist, of unknown age, mapped as
the Pampa schist (Dibblee, 1953) and named after
Pampa Peak, is dark gray and prominently foliated
parallel to bedding. It is a biotite-quartz-feldspar schist
similar to that of the Kernville series. The most south-
westerly exposures of the schist in Cottonwood Canyon
contain numerous large crystals of andalusite (chiasto-
lite) elongated parallel to foliation planes. The Pampa
schist is of sedimentary origin, having been metamor-
phosed from clay shale.

Kernville Series. The linear inclusions of metasedi-
ments exposed in the Sierra Nevada from Walker Basin
southward to Keene and again on Bear Mountain ridge
and Brite Valley were mapped as the Kernville series,
because they are similar to the Kernville series mapped

(23)

Earthquakes in Kern County, 1952

[Bull. 171

Figure 1.

Mole-track starii exteiuHiiK northeastward, close to trace of White Wolf fault. View
northeast toward Bear Mountain. Photo hy Lauren A. Wright.

by Miller and Webb (1940, pp. 349-353, map) in the
Kernville quadrangle. The Kernville series is a sequence
of metamorphosed marine sediments composed predomi-
nantly of biotite-feldspar schist with prominent platy
foliation parallel to beddiiifj. Interbedded with the
schist are layers of gra.v-white quartzite and fine tex-
tured gray to white limestone. The maximum exposed
thickness of the Kernville series within the mapped area
is about 6000 feet, but the original thickness was no
doubt many times that amount. The age of the Kern-
ville series is unknown, as no diagnostic fossils have
been found in it.

Limestone of Tehachapi Range. On the southeast
slope of the Tehachapi range are several isolated ex-
posures, within and south of the Garlock fault zone, of
metamorphic blue-gray to white limestone containing
minor interbeds of gray-white quartzite, lime-silicate
hornfels and black schistose biotite hornfels. This forma-
tion is probably the same as the Bean Canyon schist-
limestone series described by Simpson (1934, pp. 381-
383) exposed 8 miles beyond the east border of the
mapped area, and may be also the equivalent of the
Kernville series north of the Garlock fault.

Schist of the San Joaqvin Valley. Schist underlies
the Tertiary sediments under a large portion of the
east side of San Joaquin Valley, being encountered in
wells throughotit the Arvin-Mountain View-Edison oil
field area, northward to the Ant Hill field, and south-
ward nearly to the White Wolf fault. This buried schist
is probably continuous with the outcrops of the Pampa
schist exposed in Cottonwood Canyon to the northeast.

The schist immediately underlying the Tertiary sedi-
ments was shattered or rendered porous bj' weathering

l)rior to deposition of the sediments; and in portions of
the Edison and Mountain View-Arvin oil fields, it is a
reservoir rock for oil derived from the Tertiary strata
where they buttress against the schist up dip. Basement
cores from many wells in and near these fields have
been examined by May and Hewitt (1948, pp. 129-158),
who have determined the character and distribution of
the schist under this part of the valley.

Diorite-Gabbro. In the western Sierra Nevada foot-
hills there are three small exposures of a dark intrusive
rock ranging from diorite-gabbro to gabbro. One of
these occurs in Kern Gorge, another in Rattlesnake
Canyon and one in the foothills east of the Rockpile.
A large exposure of gabbro crops out east of Pastoria
Canyon in the Tehachapi Mountains. These exposures
are composed of dark-gray to nearly black medium-
textured, e(|uigi-aiiular rock composed almost entirely of
calcic plagioclase feldspar and hornblende.

Quartz Diorite. The granitic rock .so extensively ex-
posed throughout the southern Sierra Nevada within
Caliente quadrangle has been determined petrographi-
cally as quartz diorite (C. W. Chesterman, in Dibblee
1953). This rock, predominant in the vicinity of the
White Wolf fault, is composed of quartz and white
orthoclase and plagioclase feldspars with plagioclase
predominating, and biotite mica and hornblende in
varying amounts, but the dark minerals seldom exceed
30 percent of the total rock mass. The rock is light to
medium-gray, depending on the amount of dark min-
erals present, is medium-textured, equigranular. Two
faeies of quartz diorite are developed in the southern
Sierra Nevada and appear to have intruded the pre-
existing rocks at different times or in diiferent modes.

Part n

Geology

In the San Joa(iiiin Valley the basement complex
underlying the Tertiary sediments is made up largely
of quartz diorite. From the foothill area between Cali-
ente Canyon and the White Wolf Ranch the quartz
diorite extends westward under the adjacent portion of
the valley wliere it has been found below the Tertiary
in all well cores. Well logs indicate the quartz diorite-
sehist contact exposed 2 miles north of Bena extends
southwestward 6 miles to the Edison oil field, through
which it curves scnithward passing .just east of Arvin
probably to the White Wolf fault. In the vicinity of
the Kern River and Bakerstield most wells that pene-
trated to the basement complex cored quartz diorite.

Granite. A nearly white massive plutonic rock
mapped as true granite by Wiese (1950, pp. 24-25) crops
out on the south side of the Tehachapi Range south of
the Garlock fault. Numerous dikes of pegmatite and
aplite cut the foliated quartz diorite, especially at or
near the borders of the massive faeies from which the
dikes may have originated. The pegmatite is composed
of very coarse textured quartz and white feldspar, with
gradations to fine textured aplite of the same minerals.
The dikes range from less than an inch to 10 feet thick,
and are especially numerous east of Caliente and in
Rattlesnake Canyon where they trend northeast parallel
to the contacts of the massive quartz diorite and dip
steeply toward it.

Hypahyssal Intrusives. Fine textured intrusive igne-
ous rocks have been cored in the basement complex by
many wells that have reached it in the Bakerstield, Edi-
son, and Mountain View areas, as reported by ilay and
Hewitt (1948, pp. 141-3). These rocks are hard, massive
to slightly schistose, light to dark gray or greenish gray,
and are composed of aplite, malchite (mierodiorite),
andesite, diorite-aplite, diorite porphyry, and lampro-
phyre.

Intrusive rhyolite of probable Cenozoie age crops out
1 mile ea.st of Keene as several small dikes up to 12 feet
thick cutting quartz diorite. The rhyolite is a dense
cream-white rock weathering tan, with small phenocrysts
of quartz and feldspar. Associated w^ith one of the rhyo-
lite dikes is a small deposit of cinnabar at the Walibu
mine.

Tertiary-Quaternary Series

The basement complex of the Sierra Nevada and
Tehachapi-San Emigdio iSIountains is unconformably
overlain by the Tertiary and Quaternary sequence of
sediments ranging in age from Eocene to Recent. Where
the series is buried under the San Joaquin Valley the
stratigraphy has been determined from numerous well
logs. The series crops out as a strip along the Sierra
Nevada foothills as far southeast as Caliente Canyon
where the outcrop section is terminated b.v the Edison
fault. Between this fault and the White Wolf fault to
the southeast the Tertiary is not exposed. South of the
latter fault the series again crops out in the Tejon Hills
and along the foothills around the southeast end of the
valley into the San Emigdio Mountains.

The Tertiary series within the mapped area is a con-
tinuous succession characterized by numerous and rapid
changes of faeies and thicknesses. In general the series
thickens from east to west, absent along the foothills
between the Edison and White Wolf faults and as thick
as 25,000 feet just north of Wheeler Ridge. This west-
erly thickening is accompanied by gradation from coarse
detrital material along the eastern margin to tine-grained
argillaceous, thin-bedded sediments as the series thickens
toward the deeper portion of the San Joaquin de-
positional basin. The general stratigraphic succession is
as summarized below.

Marine sediments of Eocene age, the Tejon formation,
overlie the basement complex and crop out only at the
southern end of the San Joaquin Valley. This formation,
which lies deeply buried under the westerly portion of
the valley and does not extend as far east as the over-
lying strata, was deposited in a sea which transgressed
from the west.

The Tejon formation and the basement complex are
overlain by the nonmarine Tecuya formation of the San
Emigdio-Tehachapi foothills, and the Walker formation
and Bealville fanglomerate of the Sierra Nevada foot-
hills, all of Oligocene (?)-lower Miocene age. These are
made up of coarse, land-laid sediments containing some
volcanic lavas and tuffs, deposited after regression of the
Eocene sea. These sediments eventually grade westward
into marine strata of the Pleito formation and its
equivalents.

Figure 2. l>etail of mole-track scarp (pressure ridge) about
4 miles due east of Arvin. Bear Mountain in background. Photo by
Lauren A. Wright.

i: I

'^"■^-^iL ,..^-v..

14 I II

■•." t*«-

I'lij 1:1 :;, I'.mkiii iniicc, .-iu^'jjfsting compression, across White
Wolf fault east of Highway 4t;6, at base of Bear Mountain east of
Arvin. Photo by Lauren A. M'right.

Earthquakes in Kern County, 1952

[Bull. 171

The Tecuya and Walker formations are overlain by
marine sandstones and clay shales of the Temblor (or
Vaqiieros) formation, lower Miocene age, deposited as
the Miocene sea transgressed eastward to the present
site of the Tehachapi and Sierra Nevada foothills. The
Temblor formation crops out in the San Emigdio Moun-
tains, and on the east margin of the San Joaquin Valley
where it is represented by the Pyramid Hill, Vedder,
and Jewett sands. Freeman silt, and Olcese sand. The
Temblor formation is followed by the Maricopa (or
Monterey) shale, a series of organic shales of middle
and upper Miocene age deposited in the Miocene sea
when it reached its maximum extent and depth. This
formation extends eastward under the valley to the
Tejon Hills and Sierra Nevada foothills where it is rep-
resented by the Round Mountain silt, Fruitvale shale,
and marginal sand lenses. Overlying the Maricopa shale
and its equivalents is the Santa Margarita sand, which
is fairly persistent throughout the mapped area and
represents the final stage of marine deposition as the
Miocene sea regressed.

All of the above Miocene marine formations grade
eastward into terrestrial facies mapped as the Bena
gravels, which crop out in the Sierra Nevada foothills,
Tejon Hills and Tehachapi foothills. There coarse detri-
tal sediments were deposited as piedmont alluvial fans
on the coastal plain bordering the Miocene sea whose
strandline persisted along the present site of the south-
eastern margin of the San Joaquin Valley.

The Miocene strata are overlain by terrestrial sedi-
ments of the Chanac and Kern River formations, of
Pliocene age, the latter ranging into Pleistocene. These
coarse alluvial sediments crop out in the lowest portions
of the Sierra Nevada foothills, Tejon Hills, Tehachapi
foothills and Wheeler Ridge, and underlie San Joaquin
Valley where the series attains a maximum thickness of
14,000 feet just north of Wheeler Ridge. The Chanac,
and perhaps the lower part of the Kern River formation,
eventually grade westward into the brackish-marine
Etchegoin formation ; the first definite marine beds ap-
pear in the outcrop section about 8 miles west of
Wheeler Ridge, beyond the mapped area.

Sumtnary of Geologic History

Pre-Cambrian (?). Accunuilation of muds, volcanic a.sh, and
minor anionnt.s of sand and lime, prolialily in an open sea, now
the gneiss and I'elona schist in the Tehachapi Mountains.

Paleozoic (f) to Jurassic (?). Accumulation of muds, sands,
limestones and some lavas of the Kernville and Panipa series under
an open sea to a tremendous thiclcness, resulting; in a very deep
burial of the lower strata and of the underlyinp strata.

Late (?) Jurassic Nei-adan Orogeny. Strata deposited during
])re-Cambrian (?) to .Tura.ssic (?) time subsided to such great
depth (10 miles or deeper) that they became subject to regional
thermodynamic metamorphism. Pre-('ambrian ( ?) formations
altered to gneiss and the Pelona schist, and Paleozoic (?) or early
Mesozoic (?) strata to schist, quartzite, marble and meta-igneous
rocks of the Pam|)a and Kernville series. At this great depth meta-
morphosed rocks were intruded by molten magmas which crystal-
lized into granitic rocks, chiefly quartz diorite.

Cretaceous. Long interval of erosion ; mountainous terrain built
up during Nevadan orogeny deeply ero<led to surface of low relief ;
westward tilt initiated, causing continued or renewed uplift of
Sierra Nevada-Great Basin region and downward tilt of area to
west under Pacific ocean.

Korenr. Continuation of last event ; Sierra Nevada probably
blocked out during this time by faidting ; downtilted area to west
submerged under waters of ocean that transgressed from west ;

deposition of sands and clays of Tejon and equivalent formations
on deeply eroded and peneplaned surface of metamorphic and
granitic basement complex.

Oliyocene (?). Continued westward tilting: Sierra Nevada
underwent renewed uplift and erosion ; eroded debris deposited
along western liase as alluvial fans of the Bealville, Walker and
Tecuya formations, causing partial regression of Eocene .sea ; local
volcanic eruptions of pumiceous ash and some basaltic lavas.

Miocene. Continued westward tilt with some renewed uplift
and erosion of ancestral Sierra Nevada ; eroded debris deposited
along western base as alluvial fans of the Hena gravel ; down-
tilted area continued to submerge under a great open sea trans-
gressing from the west, with ancestral shoreline along base of
Sierra Nevada uplift fnmi site of mouth of Kern Gorge southward
through Kdi.son-Mountaiu View-Arvin oil fields, Tejon Hills, and
across Tehachapi Mountains; sands deposited in shallow waters
near shore, clays and siliceous muds in deeper waters farther to
west. Sea transgressed from west in early Jliocene time, regressed
at end of Miocene; Edison fault developed during Miocene deposi-
tion, with uplift of south block.

Pliocene. Continued westward tilt; renewed uplift and erosion
of Sierra Nevada ; Tehachapi and San P^niigdio ranges formed by
copipressive uplift against Garlock and San Andreas faidts respec-
tively ; San .loaiinin embaymeut thus formed with dejiosition of
Etchegoin formation at southeast extremity of this lingering em-
baymeut ; debris eroded from mountains deposited as alluvial sedi-
ments of Chanac and Kern River series, thus filling embayment
to form San ,Ioa(|uin Valley; major and some minor faults in
area may have been initiated.

Pleistocene-Recent. Recurrent uplift of Sierra Nevada, involv-
ing adjacent foothill area, causing partial uplift and erosion of
Tertiary sediments deposited along western base ; development of
Kern River fault and numerous other faults in foothill area and
eastern San .Joaquin Valley ; Kern Canyon fault ; uplift of Breck-
enridge Mountain block on Breckenridge fault and consequent filling
of upiier Walker Basin Canyon to form Walker Basin ; develop-
ment of Garlock and White Wolf faults with uplift of Bear
Mountain, Bear-Brite-Cummings-Tehachapi Valley and Tejon Hills
area as a single block between them ; recurrent compressive uplift
of Tehacbaiii-San Emigdio Range against Garlock and San Andreas
faults; compressive northward movement and elevation of San
Emigdio uplift on I'leito thrust fault and rise of Wheeler Ridge
anticline and foothills in front of San Emigdio overthrust ; material
derived from erosion of all elevated areas deposited as alluvial fill
in San .Toaquin A'alley.

GEOLOGIC STRUCTURE

Structural Setting of the Southeastern San Joaquin
VaUeij Region. The southern Sierra-Great Valley prov-
ince of California constitutes a regional structural block
about 120 miles wide and roughly 600 miles long, made
up of a complex of cr.vstalline metamorphic and intrusive
granitic rocks stabilized to a comparatively rigid mass
during the great Nevadan orogeny at the end of Jurassic
time. During Cretaceous and Cenozoic time this huge
structural block was tilted westward almost continuously,
with the eastern portion elevated to form tlie Sierra
Nevada, and the western portion tipped downward to
form the Great Valley. As this block was tilted, material
eroded from the rising Sierra Nevada was carried west-
ward and deposited to form an enormous thickness of
Cretaceous-Cenozoic sediments in the sinking Great
Valley area which was submerged under marine waters
until late Tertiary time when it became filled with
sediments. Westerly tilting caused progressively more
rapid subsidence of the west side of the Great Valley
so that the basement complex is progressivelj' more
deeply buried from east to west, being buried many
miles deep on the western margin.

The Sierra Nevada has been uplifted as a huge west-
tilted block on the nortli-trending Sierra Nevada fault
zone along its eastern base. In this zone faulting is of
the normal type with fault planes dipping steeply east-

Part I J

Geology

ward. The northern and central portions are made up
"f several en echelon north-trenclinfr faults, while the
southern portion is a siiififle fault alon;,' which the moun-
tain hlock was elevated to great hoif;hts to form the
imposing front of the hijrh Sierra. The southern Sierra
Nevada fault is paralleled about lo miles west by the
Kern Canyon-Breekenridge faidt along which the west-
ern portion of the southern Sierra Nevada was elevated.

The pivotal area of the continuously westward tilted
Sierra Nevada-Great Valley block follows the present
margin between the Sierra Nevada and the Great Valle,v,
as topographic and stratigraphic evidence indicates this
marginal area to have been neither elevated nor de-
pressed to any great extent during Cretaceous or Ceno-
zoie time. During that long interval of time this 600-
mile long pivotal area has been remarkabl.v stable and
free from tectonic movements. Only in the extreme
southeastern portion in the vicinity of Bakersfield have
tectonic disturbances occurred and these have been in
the form of faulting. The largest of these faidts is the
"White Wolf fault trending northeast directly across
the pivotal margin between the Sierra Nevada and the
San Joacjuin Valley. The area southeast of this faidt was
elevated as a block to form the great westward protru-
sion of the Sierra Nevada or the Bear Jlountain uplift.
Geologic and seismic evidence indicate this i'atilt to be
a southeast-dipping reverse fault with lesser left lateral
movement.

The Sierra Nevada-Great Valley province is termi-
nated on the southeast by the northeast-trending Tehach-
api Range and uplift, traversed along its crestal portion
by the Garlock fault. This is a master fault extending
150 miles north of east from its juncture at Lebec with
the San Andreas fault, and separates the Sierra Nevada
and Basin-Range provinces on the north from the Mojave
Desert province on the south. This fault is a great active
shear zone of a type similar to the San Andreas, although
movement on the Garlock has been left lateral, with the
north block having moved westward relative to the south
block.

The Tehachapi Range merges westward with the San
Emigdio Range which trends north of west and forms
the southernmost of the inner Coast Ranges bordering
the San Joaquin Valley on the southwest. These ranges
have been formed by compressive uplifts recurring dur-
ing Cenozoie time along or near the northwest trending
San Andreas fault, largest and most vigorously active
crustal break in California. Movement ou this 600-mile-
long vertical shear zone has been largely horizontal,
right lateral, with the southwestern block having moved
northwest relative to the northeastern block. This fault
has been recurrentlj% if not continuously, active through-
out Cenozoie time, and cumulative right lateral displace-
ment has amounted to many tens of miles. The horizontal
shear movement has been accompanied by crustal short-
ening, which in the San Emigdio Range on the northeast
side of this great shear zone has been so severe that the
once deeply buried rigid basement complex has been
shattered and squeezed up to form the high crest of the
mountains. The overlying thick Tertiary series has been
folded and thrust northward toward the San Joaquin
Valley on the Pleito fault along the foothills to form
the San Emigdio overthrust.

It is noteworthy that the San Emigdio overthrust lies
north of the intersection of the San Andreas and Gar-
lock faults, two master shear zones with oi)posite hori-
zontal movement. There is suggestive evidence that the
White Wolf fault may extend at depth southwest to the
San Andreas, and that it may have some left lateral
component of movement like the Garlock fault. The
White Wolf and Pleito faults are probably genetically
related to the Garlock and San Andreas shear faults and
not to any of the faults to the noi'th.

The White Wolf Fault

Geologic Evidence. The steep northwest slope of the
high westward protrusion of the Sierra Nevada between
Tehachapi Canyon and the Tejon Hills has long been
recognized as the scarp of a major fault. Such a fault
was first recognized by Lawson (1906), and was mapped
by Hoots (1930. p. 314), as the White Wolf fault, named
after the White Wolf Ranch through which it pa.sses.

The White Wolf fault trends from lower Tehachapi
Canyon S 50° W for 17 miles along the base of the
steep northwest slope of Bear Mountain to Comanche
Point, and probably extends at least an equal distance
across the San Joaquin Valley toward Wheeler Ridge.
This fault is nowhere clearly exposed and is thereby
difficult to trace as the northeastern portion is within
quartz diorite. the central portion is covered by numer-
ous landslides from the elevated block, and the south-
western portion is concealed by alluvium. Prior to July
21. 1952. practically nothing was known about this
faidt. Knowledge of its existence was based entirely on
topography and was later substantiated by deep drilling
in the valley area north of the fault and at Comanche
Point. The steep, abrupt slope of Bear Mountain indi-
cates the fault to have been active in very late Pleisto-
cene and Recent time. Its genei-ally straight base line
might lead to the interpretation that the White Wolf
fault is a normal fault, but the presence of numerous
landslides and earth flows from the elevated mountain
block suggests the fault to be of the reverse or thrust
t.vpe. Left lateral movement on this fault might be in-
ferred from its parallel trend with the left lateral Gar-
lock fault, but this is not conclusively indicated by topo-
graphic or stratigraphic evidence.

Surface Ruptures Developed on July 21, 1952. The
surface ruptures developed in the ground along the
course of the White Wolf fault during the earthquake
of July 21, 1952, if these represent the actual move-
ment on the fault, serve to determine its exact location
at the surface and indicate its probable direction of dip
and nature of movement. Surface cracks formed along
almost the entire 17-mile known course of the White
Wolf fault, and most of these occurred on or very near
its trace. The fracturing occurred mainly along the
northeastern, central and southwestern portions of the
fault, with gaps in between. The type of rupturing in
each portion varies greatly and there is some doubt as
to whether these all represent actual tectonic fault
movements which caused earthquakes or gravitational
earth settling movements which resulted from the earth-
quakes. For reasons stated in the following paragraphs,
these large cracks along the White Wolf fault are be-
lieved to be primary features of fault movement on
July 21, 1952, as expressed at the surface.

Earthquakes in Kern County, 1952

[Bull. 171

Fk)1!re 4. Northwestern slope of Bear Mountain, showing landslide toiiography. Two earth(|iiake craoks, roughly at right angles, are
near center of photo. The more prominent crack trends northeast and is close to the trace of White Wolf fault. Official iihotoyniph. I . .S. Air
Forces, Edwards Air Force Base, Edwards, California.

In areas upslope from the White Wolf fault there are
many small ruptures on the steep slopes of Bear Moun-
tain and on both sides of Sycamore Canyon up to the
4,000 foot contour. These ruptures are 10 to about 200
feet long and either follow a contour or in most cases
are concave down slope. They are always gapino- and dip
steeply downhill with the hanging block having always
slid downhill one or several feet. One of these on the
south slope of Bear Mountain slid down as much as 30
feet. These are small shallow landslide features which
developed only in the soil or in weathered or shattered
quartz diorite on steep slopes, and were produced by
gravitational settling of this material by lurching re-
sulting from the main shock; they are therefore not
faults.

The ruptures developed along the trace of the White
Wolf fault are the largest and most extensive, and are
believed to be true faults. The greatest and mo.st con-
tinuous zone of fracturing exists along the central 5
miles of the White Wolf fault or that portion following
the base of the steep 5,000-foot-high granitic scarp of
Bear Mountain beginning at a point 4 miles east of
Arvin and extending contiiuiously for 3 miles northeast-
ward, then intermittently for another 2 miles to the
canyon south of the White Wolf ranch house. The entire
mountain block on the southeast side of this zone of
fracture was elevated 1 or 2 feet, and thrust toward the
northwest. The fracturing along this 5-mile portion of

the White Wolf fault is consequently characterized by
thrust fault scarplets usually facing northwest, and as-
sociated pressure ritlges or mole tracks. In most places
the upthrust block formed a single scarplet a foot or
two high and traceable for several Iniiulred feet. All the
scarplets along this portion of the fault were miniature
overthrusts with the plane of movement dipping south-
east at low angles and with displacements toward the
northwest or west of north. There was no evidence of
obli(|ue movement except where the scarplet deviated
from the usual northeast trend, in which case the move-
ment was always toward the northwest. In places where
a scarplet was not developed, the same amount of
shortening was taken up by a series of parallel pressure
ridges or mole tracks. The hard dry soil was always
broken up into small irregular blocks along the scarplets
and ridges. However, in some places where they trend
obliquely to the normal northeast trend, oblique tension
fissures trending northwestward were developed along
each scarplet. The scarplets and bucklings pass around
spurs and extend up gullies or small canyons where they
indicate an attitude of the plane of movement of about
5° to 20° southeastward toward the moimtain mass with
a probable average dip of about 15°. Throughout most
of its extent, the zone of scarplets and bucklings follows
the exact base of the steep mountain front along the
contact between quartz diorite above and alluvium
below. Only at the west end do the scarplets appear to

Part n

Geology

(lie out into alluvinni. Init mappintr iudioatcs tliat here
also the elevated southeastern block is uiiderlaiu either
on the surface or at shallow depths by ipiartz diorite.

It seems clear that a thrust fault is indicated by the
o-mile-lons zone of scarplets and pressure ridges at the
base of the granitic scarp of Bear Mountain. Crustal
shorteniuf,' is definitely indicated, not only by the pres-
sure ridtres and miniature overthrusts, but in one in-
stance by a fence crossing this zone of buckling in which
several posts were pushed several inches toward each
other, leaving the wires sagging. There is some doubt as
to whether this zone of surface thrusting is an actual
thrust movement on the White Wolf fault, or whether
it is the result of landslide or earthflow movement of
shattered material from the steep slope of Bear Moun-
tain, as it was interpreted by Buwalda (1952, p. 5).
Support for the latter interpretation is found in the
generally shattered condition of the quartz diorite and
resulting landslide topography of the steep northwest
slope of Bear ^Mountain, and by the occurrence of this
zone of thrusting only at the base of this high, steep
slope. However, the following evidence seems to indicate
thrust rather than landslide movement :

(1) Xot all of the lower northwest slope of Bear Mountain is
characterized by landslide to|iography, but much of this sloi*
rises abruptly from the base and is intact.

(2) There is no large scale rupturing or separation of material
upslope on Bear Mo\intain from the zone of buckling as would
be exi)ected if the buckling at the base of the steep slope
resulted from downward movement of material from upslope.

(3| The (luartz diorite is shattered only on the lower slopes of
Bear Mountain and is generally intact elsewhere.

(4) The plane of movement is not horizontal nor does it dip
downsloiie, but dips into the mountain throughout its course

at an average angle of about 15°.

(5) The White Wolf fault at the base of Bear Mountain has all
the characteristics of a reverse or thrust fault, similar to the
Pleito thrust, and not of a normal fault, as its course is ir-
regular— the bedrock immediately above it is highly shattered
and the scarp is characterized by numerous landslides as is
always true of thrust fault scarps.

(6) The July 21 earthiiuake failed to move any of the large land-
slides on the steep northwest slope of Bear Mountain, a fact
indicating that even large landslides are superficial features
developed only during or following periods of heavy rainfall.

It is concluded that the zone of surface thrusting at
the base of Bear ^Mountain is most likely the result of
actual thrust movement on this portion of the White
Wolf fault along which the mountain block was thrust
upward and toward the San Joaquin Valley about 3
feet on July 21, 1952.

The zone of rupturing along the northeastern 6 miles
of the White Wolf fault between the White Wolf ranch
house and the railroad tunnels east of Bealville is char-
acterized by ruptures quite different from those of the
central portion. The fractures of the northeast portion
did not follow the exact trace of the fault, but occurred
as vertical ruptures en echelon along and oblique to it
with a more northerly trend. Between the White Wolf
ranch house and the Bealville road there are two or
three series of north-trending vertical fractures 1 mile
to 3 miles long. These fractures developed as small pres-
sure ridges or ruptures in the soil indicating left lateral
movement (west block moved south relative to east
block) in every case, with fences and highways offset a
foot or two. These fractures did not form scarplets ex-
cept in irregular topography, in which the scarplets

faced either direction, depending on the direction of
slope, and always indicated left lateral displacement.
The consistent left lateral offsets on these ruptures show
them to be true fault cracks formed by tectonic move-
ments rather than landslide cracks formed by earth
settling movements.

Between the Bealville road-U. S. Highway 466 inter-
section and Tehachapi Creek, the White Wolf fault is
within quartz diorite and the steep northwest-facing
scarp gradually disappears rendering this portion of
the fault difificult to locate. Its position is believed de-
termined by an alignment of saddles and small canyons
extending north of east from the road intersection, and
also by the highly sheared, crushed and shattered con-
dition of the quartz diorite on the north side of this
alignment. In the July 21 earthquake fractures devel-
oped in the shattered quartz diorite immediately north
of this supposed fault trace. The fractures between the
road intersection and Southern Pacific Railroad Tunnel
5 trended generally northeast and showed evidence of
left-lateral displacement and strong compressive move-
ment indicated by contortion of the rails into S-shaped
curves and shortening at tunnel 3 east of Bealville.
On the south slope and crestal portion of the large hill
through which tunnel 5 passes there were four parallel
fractures trending west of north for half a mile and
dipping steeply north (see photographs. Part III, Con-
tribution 6). On each of these fractures the northern
or uphill side slipped down so that movement on each
one formed a gaping fissure up to 6 inches wide and a
scarplet up to 2 feet high facing northward upslope.
There was no evidence of lateral slip. These are prob-
ably minor tensional cross faults to the main course of
the White Wolf fault, or might be the result of gravi-
tational settling or lurching of shattered quartz diorite
northward toward Tehachapi Canyon.

There is no topographic or structural evidence that
the White Wolf fault extends beyond Tehachapi Creek
although one small northeast-trending crack formed on
the lower northea.st slope of the canyon. One large north-
trending crack extending about 1000 feet developed
across the Tehachapi-Caliente divide, along the contact
between quartz diorite and the Bealville conglomerate.
This crack dips steeply west and formed a 3-foot scarp
facing west.

In Caliente Canyon many very small north-trending
cracks formed across the road in shattered quartz diorite
for about a mile east of Harper Canyon. These appear to
be fractures along jointing in the quartz diorite with no
movement indicated.

On the southwestern portion of the White Wolf fault
fractures developed along the northwestern base of the
hills at Comanche Point and again at the mouth of Little
Sycamore Canyon. Others developed about a mile south
of the White Wolf fault near the mouth of Comanche
Creek and in the foothills to the east. All these fractures
were vertical or steep and produced northwest-facing
scarplets up to a foot high.

At Comanche Point the alignment of scarplets fol-
lowed the northwest base of the hills for a mile and the
scarplets were vertical and up to 8 inches high. Some
showed a small lateral component of movement of several
inches.

Earthquakes in Kern County, 1952

[Bull. 171

Figure 5. !>i'vurf i-i-.ickiiif; in water-satu-
rated low terrace in lower Comanche Creek ;
apparently east of White Wolf fault. Photo hy
(lonloii li. Oakeshott.

The SL-arplets at the mouth of Comanche Canyon and
in the foothills to the east were vertical and up to 8
inches hijrh with no lateral motion indicated. These
cracks did not seem to follow any definite alifi:nment, but
tended to be parallel to the northeast strike of the Cha-
nac formation here. They may have been secondary
effects produced by settling: movements. At the springs
near the mouth of Comanche Canyon, the alluvium was
broken by numerous cracks, some of which produced
small mud volcanoes of fine .sand. These cracks could
not be traced into the hills on either side of the canyon;
they were obviously produced by lurchinn: of the water
soaked alluvium.

At the mouth of Little Sycamore Canyon a continuous
zone of ruptiiriufj up to 70 feet wide was strongly de-
veloped on the White Wolf fault for a mile and a half
along the base of the foothills. On the side-hill to the
southwest a series of pressure ridges formed in the soil.
From the mouth of the can.yon northeastward the frac-
ture zone followed the edge of the valley alluvium and
was developed as a series of northwest-facing scarplets
up to a foot high indicating an overall uplift of the
southeastern block of 1 J or 2 feet. The fractures were
vertical or dipped steeply southeast, and nearly all
showed left-lateral offsets up to 8 inches. In places where
no defined scarplets were formed, especially where the
fracture zone was dying out northeastward, the hard
dry soil was broken by a series of en echelon north-
trending gaping ruptures. These were tension ruptures

produced by left lateral motion (southeast block moved
northeastward) along this portion of the White Wolf
fault.

The fracturing of both Comanche Point and at the
mouth of Little Sycamore Canyon was no doubt devel-
oped along the trace of the White Wolf fault, and indi-
cated oblique movement on this portioii along which the
southea.stern block was elevated up to 2 feet and dis-
placed northeastward about half a foot. There is some
doubt as to whether these fractures represent actual
movement on the fault or gravitational settling of the
thick valley alluvium, but the generally consistent left
lateral displacement on so many of these fractures is
difficult to explain by movements other than tectonic or
fault movements.

The ruptures along the course of the White Wolf fault
indicate a displacement of as much as 3 feet on July
21, on which the southeastern block was elevated and
shoved laterally to the northeast. However the displace-
ment at depth at and near the focus was probably much
greater than that at the surface, as indicated by the
fractures, as the displacement probably decreased up-
ward and in many places failed to reach the surface.
In the valley area southwest of Comanche Point past
which the fault is believed to extend, the displacement
may have been completely absorbed by the thick sedi-
mentary fill ; this could account for the total absence of
fractures there.

Total Displacniicnf. The amount of total vertical dis-
placement on the White Wolf fault is indicated by the
height of the Bear Mountain scarp on the elevated block
plus the depth to the basement complex of the relatively
depressed block, or the difference of the depth to the
basement complex of each block where covered by Ter-
tiary sediments. Depth to the ba.sement complex under
the alluviated valley floor is indicated by subsurface
contours on the geologic map, based on points on the
top of the basement complex as encountered in all wells
that reached it. From a study of the geologic map, it
may be seen that the total upward displacement of the
southeast block of the White Wolf fault increases rap-
idly from none at Tehaehapi Canyon to 5,000 feet or
more at the base of Bear Mountain, and 10,000 feet
between the mouth of Little Sycamore Canyon and Co-
manche Point. Southwest of Comanche Point the dis-
placement is unknown, but it probably decreases.

Some left lateral displacement on the White Wolf
fault is iiulicated by movement on the faidt fractures
as mentioned. However, the total overall lateral dis-
placement must be small, probably not over 2000 feet.
The easterly pinchout of the Santa Margarita sand is in
about the same position on either side of the fault and
therefore is not appreciably offset. There are no defi-
nitely offset streams along the course of the fault.

Dip of Fault Plane. The dip of the White Wolf
fault plane is to the southeast as indicated by both the
surface ruptures and by the physiographic expression
of the landslide-covered Bear Mountain escarpment.
The amount, as indicated by the ruptures, ranges from
10° to 90°. The ruptures indicate a low dip for the
central portion and a high dip for the southwestern and
northeastern portions. Perhajw the overall dip may best
be indicated by the location of the epicenter — if accu-

Part I]

Geology

Figure 6. Mud volcanoes alonfc cracks in water-saturated low terrace in lower Comanche
Creek ; apparently east of White Wolf fault. Photo by Gordon B. Oakeshott.

rately determined. Accordinpr to St. Amaud (oral com-
numieatioii, November 1952), one important epicenter
was located almost directly under Bear Mountain at
about 12 miles below sea level. Projecting this position
up to the nearest surface trace would determine a dip
of about 70° southeast.

Prom the foregoing it appears that the only surface
fractures that represent the true dip of the White Wolf
fault are those at and near tlie mouth of Little Syca-
more Canyon. The low angle thrust feature at the base
of Bear Mountain must then be a local flattening of the
fault at the surface where the elevated mountain block
partially overrode the San Joaquin Valley area. The
north-trending left lateral fault cracks along the north-
eastern portion of the White Wolf fault probably branch
off from the main fault below the surface and were pro-
duced by upward and northeastward movement of the
southeastern block.

Type of Fault and Movement. The foregoing data
indicate the White Wolf fault to be a high angle reverse
fault dipping southeast along which the southeastern
block was elevated to a maximum displacement of some
10,000 feet and displaced a much lesser distance to the
northeast — relative to the stationary northwestern
block. The low-dipping thrust fault rupture along the
central portion of the fault indicates that the north-
western or footwall block is stationary and tliat the
southeastern block was actively elevated and thrust
northwestward. This is further suggested by the inten-
sity of the earthquake of July 21, 1952, which was more
violent in the area southeast of tlie White Wolf fault
than in the area to the northwest.

Possible Northeastivard Extension. There is neither
physiographic nor geologic evidence that tlie White Wolf
fault extends northeast of Tehachapi Canyon and there
is no evidence that the White Wolf fault ties to the
Breckenridge Mountain fault. However, between the
Tehachapi and Caliente Canyons several isolated rup-
tures trending nearly northward did develop in shat-
tered quartz diorite. These may have formed along one

or several north-trending branches of the White W^olf
fault that might extend at depth as far northward as
Caliente Canyon.

Possible Southwest Extension. The extent of the
White Wolf fault southwesterly from Comanche Point
is unknown as there is no direct surface indication of
this fault beyond that point, and no surface ruptures
were formed during the earthquake of July 21, 1952.
The White Wolf fault apparently does not reach the
surface anywhere southwest of Comanche Point. How-
ever, stratigraphic, structural, subsurface, geophysical
and seismic evidence indicate or suggest that the White
Wolf fault does extend southwestward across the San
Joaquin Valley and at depth under Wheeler Ridge and
the San Emigdio foothills, possibly to the San Andreas
fault. The exact location of this buried portion of the
White Wolf fault is as yet unknown, but available evi-
dence indicates it to maintain the same S 50° W trend
as does the exposed portion between Tehachapi Canyon
and Comanche Point.

Evidence that the White Wolf fault extends southwest
from Comanche Point across the southeastern San
Joaquin Valley to Wheeler Ridge is~(l) the 10,000
foot displacement at Comanche Point, indicating the
fault to extend far beyond that point; (2) the abrupt
change of the water table at the supposed trace of the
fault across the valley; (3) differences in depth of geo-
physical reflections on either side of this buried fault;
and, (4) the much greater depth to the base of the
Plio-Pleistocene continental sediments in the valley area
on the northwest side of the buried fault as encountered
in deep wells in which the maximum drilled depth to
this horizon is 14,000 feet on the northwest side of the
fault and 4,000 feet on the southeast side. Although no
well has reached the basement complex in the deeper
portion of the valley area on either side of the fault,
the marine formations underlying the continental Plio-
cene strata are consequently much more deeply buried
under the valley area on the northwest side of the sup-
posed extension of the White Wolf fault than on the
southeast side. The great difference in thickness of the

Earthquakes in Kern County, 1952

[Bull. 171

Plio-Plcistocene continental series on opposite sides of
this fault indicates it to have been active during deposi-
tion of those sediments.

Evidence that the White Wolf fanlt extends under
Wheeler Ridge is as follows: (1) seismic — the occur-
rence of the main epicenter (St. Amand and Buwalda,
1953) of the July 21 earthquake, under the southwest-
ern portion of the Wheeler Ridge anticline (latitude
35° 00', longitude 119° 02'), at a depth of about 10
miles indicates that the White Wolf fault plane must
pass through that point; (2) stratigraphic — the White
Wolf fault probably passes under the northwestern
portion of the Wheeler Ridge anticlinal uplift as the
thickness of the Plio-Pleistocene series is about 4,000
feet on this structure, or about the same as in the
adjacent valley area to the east, while it is over 12,000
feet thick in the valley area to the north ; also a well
drilled at the west portion of the Wheeler Ridire anti-
cline reached the basement complex at a depth of about
12,000 feet while the basement complex is probably
more than twice as deep in the valley area to the north ;
(3) structural — faulting under the north flank of the
Wheeler Ridge anticline is suggested by its steep dip,
and also by the occurrence of southward-dipping minor
reverse faults under this asymmetric fold as encoun-
tered in deep drilling; faulting under the southwest-
ward-plunging portion of this anticline is suggested by
the northeast-trending alignment of small sharp east-
trending subsidiary folds, which appear to be the result
of left lateral movement on the White Wolf fanlt, or
one aligned with it, at depth in the basement complex.

Evidence that the White Wolf fault may extend from
Wheeler Ridge southwestward under the San Emigdio
foothills at depth is suggested by the sharp upturning
and intense folding of the Cenozoic sediments along these
foothills. This disturbed zone is joined by the Pleito
fault from the southeast, which here curves south of
west to follow this zone of disturbance for some 12 miles
nearly to the San Andreas fault, as mapped by Hoots
(1930, map). Along this trend the Pleito fault steepens
to over 50° as indicated by a well drilled in Pleito
Canyon, and the basement complex is brought to the sur-
face on the elevated southern block west of San Emigdio
Canyon. This portion of the Pleito fault and the ad-
jacent zone of sharp folding on the footwall block are
aligned with the White Wolf fault, suggesting this zone
of disturbance to have formed along or over the deeply
buried White Wolf fault zone in the basement complex
below.

While detailed mapping indicates that the Pleito fault
zone does not extend to the San Andreas fault, never-
theless a distinct bend or curve concave northeast is
developed in the San Andreas where the Pleito fault, or
the underlying White Wolf fault zone, would intersect
it if projected to it. Northwest from that point the San
Andreas fault trends consistently N 45° W, and to the
southeast it trends N 60° W. Although this bend is not
sharp it is noteworthy in being the greatest deviation
of trend in the San Andreas and suggests that this rift
zone is intersected by the White Wolf fault at depth,
that the bend may have resulted from left lateral move-
ment on the buried White Wolf fault, and that the Pleito

Fidi UK 7. Cracks and soarplets in White Wolf fault zone,
Tejon Hills, just northeast of Comanche Creek. I'hoto by Gordon
B. Oakeshott.

fault zone developed partly from uplift of the south-
eastern block on the White Wolf fault below.

Age of the White Wolf Fault. The White Wolf fault
appears to have been most active during Pleistocene and
Recent times. It may have been active during most, if
not all, of Pliocene time as indicated by the much greater
thickness of the Pliocene sediments of the San Joaquin
Valley throughout the northwest block as compared to
the southeast block as encountered in deep drilling. The
fault may have been initiated in Miocene time, although
there is as yet no definite evidence.

Regional Tectonics

The tectonic movements active in this area are the
result of constant regional strain in this part of the
earth's crust. The mapped area lies at the juncture of
three great phj-siographic provinces — the Sierra Nevada,
Great Valley, and Coast Ranges provinces. Adjacent
ones are the Basin Ranges province to the east, Mojave
Desert province to the southeast, and Transverse Ranges
province to the south. Each one of these physiographic
provinces is also a tectonic province, characterized by
a well defined strain pattern, so that within the mapped
area several related strain patterns exist.

Sierra Nevada-Great Valley Provinces. The Sierra
Nevada-Great Valley province is made up of crystal-
line rocks stabilized during the Jurassic Nevadan orog-

Part II

Geology

t'liy to a comparatively rifrid, coinpact mass and is over-
lain by a >rt>nerally little (iisturbed Cretaceous-Ceiiozoie
sedimentary series under the (ireat Valley. This large,
rijrid se<rment of the earth's erust has resisted tectonic
movements; only in the extreme southern portion, where
teotonic movements have been more severe, has it yielded
by faulting.

The southern Sierra Nevada, bounded on the east by
the Sierra Nevada normal fault zone, is partly broken
into two major north-trending blocks by the normal
Kern Canyon-Breckenridge fault zone.

Southeastern San Joaquin VaUey. The mountainous
areas surrounding the southeastern San Joaciuin Valley
are undergoing active uplift. These are rising portions
of the earth's crust caused by constant, deep seated
compressive and shear movements active throughout
Quaternary time on the San Andreas, Garlock, and re-
lated faults. Earthquake-producing displacements have
occurred on these faults several times in a century, and
an earthquake produced by slippage on one may set off
movement on another. This apparently happened during
the recent earthquake in which the movement on the
White Wolf fault that produced the shock of July 21,
1952, may have set off movement on a minor buried
fault southeast of Bakersfield that caused the Bakers-
field earth(iuake of August 22. 1952. It is also possible
that many of the aftershocks of the first earthquake may
have been caused by movements on minor faults in the
vicinity of the White Wolf fault.

The southern Sierra Nevada foothills and eastern San
Joaquin Valley northward from the White Wolf fault
are cut by many parallel faults trending northwest to
north. In the area south of the Kern River the faults
trend generally northwest, and north of the river, tend
to swing north. They trend due north in the Kern Front
oil field area north of Bakersfield. This makes a broadly
arcuate pattern. The great majority of these faults are
of normal tj'pe although some are probably vertical or
even steep reverse. Many of the northeast-dipping nor-
mal faults bound southwest-tilted blocks as indicated by
the steepened southwesterly dip of the Tertiary sedi-
ments involved. These faults are of the same type as the
major faults of the Sierra Nevada and displacements are
probably mostly if not entirely vertical, although there
are evidences of some lateral movements.

The southern Sierra Nevada is broken b.v only two
widely spaced major normal faults — the Sierra Nevada
and Kern Canyon-Breckenridge ; while the adjoining
foothill area is broken b.v moderately spaced normal
faults of moderate displacement, the largest being the
Kern River and Edison faults. The eastern margin of
the adjacent San Joaquin Valley is broken by faults
more closely spaced and with very small vertical dis-
placements. From northeast to southwest the faults
become progressively more closely spaced and their ver-
tical displacements appear to decrease outward into the
San Joaquin Valley.

The extreme .southeastern portion of the Sierra
Nevada-San Joaquin Valley provinces was elevated as
a block on the northeast-trending White Wolf fault to
form the Bear ilountain-Tejon Hills uplift. The
Tehachapi Mountains are in part a compressive uplift
formed against the northeast-trending Garlock fault, a

master shear zone separating the Sierra Nevada from the
Mojave Desert province to the southeast. Nearly all of
the movement on the Garlock fault has been left' lateral,
along which the Mojave Desert block has moved rela-
tively northeastward. This would indicate a great north-
east-southwest counterclockwise torsional stress. The
Garlock fault took up nearly all this stress between these
two provinces of crystalline rocks. However, a small part
of this stress was taken up on the White Wolf fault as
indicated by the small left lateral component of move-
ment on it.

The area between the "White Wolf and Garlock faults
has undergone some northwest-southeast crustal short-
ening, as indicated by the northerly or northwesterly
movement of the Bear ^Mountain uplift on the southeast-
ward-dipping White Wolf fault, by east-west folding in
the Tejon Hills, and by the compressive uplift of the
Tehachapi Mountains against the Garlock shear zone.
This would suggest a north-south compressive stress.

The pattern of slightly to moderately tilted north-
northwest-trending fault blocks is similar to that of the
Basin Ranges province so that both areas are apparently
under the same stress. However, too little is known to
determine what stress formed this pattern. The normal
faults indicate, at least near the surface, an east-west
tensional stress yet the mountain blocks seem to be actu-
ally rising as if heaved up from below while some of the
valleys (such as Death Valley) are apparently sinking.
The most plausible hypothesis that can be offered is that
the entire combined Sierra Nevada-Basin Ranges prov-
ince constitutes a thick segment of the earth's crust
which was arched upward probably by a very deep-
seated east-west compressive stress; and that the upper
portion of this thick segment became broken into north-
south trending blocks, some of which failed to rise, or
even sank. During this arching process, the western
margin became compressed downward to form the Great
Valley. In the eastern San Joaquin Valley, the arcuate
pattern of faults with the trends gradually swinging
from south to southeastward suggests that southward
the subterranean stress was directed progressively more
from an east-west to a northeast-southwest direction.

The extreme southeastern portion of the Sierra
Nevada-Great Valley province is thus affected by three
stresses: (1), a deep subterranean east-west and or
northeast-southwest compressive ( ?) stress; (2), a north-
east-southwest counterclockwise torsional stress; and
(3), a relativel}- shallow north-south compressive stress.

Coast and Transverse Ranges Provinces. In the San
Emigdio Mountains, the strain pattern is one of exten-
sive north-south crustal shortening as indicated by north-
ward movement of the San Emigdio uplift on the south-
ward-dipping Pleito thrust fault and by the strongly
compressed folds in Cenozoic strata with axes trending
slightly north of west. The crustal shortening developed
in the Tehachapi Mountains progressively increases west-
ward into the San Emigdio Range as indicated by
the increasing amount of movement on the Pleito fault
and increasing deformation of the Cenozoic strata west-
ward. This pattern is obviously the result of a severe
compressive stress directed from the south or slightly
west of south, and is progressively more intense west of
that active in the Tehachapi Mountains.

Earthquakes in Kern County, 1952

[Bull. 171

Flol'RE 8. View west toward Jones ranch house, showint; cracks
in alluvium. These cracks are not parallel to the fault trace but
are in an area of e.xtensive lurch crackinj; on the valley tloor. At
Edison Road about 4 miles southwest of Arvin. Photo by Lairreiice
W. Chiisteen.

The San Emig:dio Range, the southeasternmost mem-
ber of the Coast Ranges province, is a compressive uplift
built up against the San Andreas fault, as are all of the
Coast Ranges adjacent to this great right lateral shear
zone. The strain pattern of the San Emigdio Range is
thus typical of that throughout the Coast Ranges prov-
ince.

In marked contrast to the rigid Sierra Nevada-Great
Valley province, the Coast and Transverse Ranges prov-
inces together are a great zone of weakness in the earth's
crust. It is an unstable zone of intense crustal shorten-
ing and shearing in which the San Andreas faidt is the
greatest single rupture.

In the Coast Ranges province the strain pattern is
composed of several faults of the San Andreas type,
that is, northwest-trending vertical or high angle shear
faults with right lateral displacements of which the San
Andreas is the largest and most active, and a series of
tightly squeezed folds trending slightly west of north-
west and some reverse or thrust faults with a simi-
lar trend. The San Andreas t.vpe faults are deep-
seated ruptures originating many miles deep in the
basement complex. Several of these faults were active
throughout Cenozoic time with cumulative right lateral
movements amounting to many tens of miles (Hill and

Dibblee, 1953, pp. 443-458). These great shear faults
must be the result of a northwest-southeast clockwise
torsional stress. The compressive folds and lesser reverse
or thrust faults are comparatively shallow structures
affecting the Cretaceous-Cenozoic strata and are most
intense and numerous adjacent to or near the great shear
faults and decrease outward away from them — as in the
San Joaquin Valley. These compressive structures must
therefore be subsidiary to the master shear t.vpe faults
and were formed by an east-northeast west-southwest
compressive force resulting in part from right lateral
horizontal drag on the shear faults and in part from
pressure and eounterpressure of the opposing fault
blocks.

In the Transverse Ranges province, the strain pattern
is basically the same as that of the Coast Ranges province
except that the folds and reverse or thrust faults trend
more nearly east and in addition there are several major
shear type of oblique slip faults trending slightly south
of west with left lateral movements similar to the Gar-
lock and White Wolf faults. This would indicate, in
addition to the stresses active in the Coast Ranges prov-
ince, that the northeast-southwest counter-clockwise tor-
sional stress active along the Garlock shear zone across
the San Andreas fault affected the Transverse Ranges
province also.

Tectonic Imjilicaiions of the Strain Patterns. Tec-
tonic implications and relationships of the San Andreas,
Garlock and other major strike-slip faults and related
structures are discussed by Hill and Dibblee (1953),
who suggest that they are genetically related and re-
sulted from an overall single regional north-south stress.

It is concluded that the White Wolf fault is genet-
ically related to the Pleito and Garlock faults and
possibly in part to the San Andreas fault, but not to
any of the faults to the north. Both the White Wolf
and Pleito faults are in part the result of compressive
stresses developed along both the Garlock and San
Andreas shear zones. This is indicated b,v the southward
dip of both faults toward the great shear zones and
squeezing of the area between these faults and the shear
zones. The White Wolf fault is believed to be closely
related to the Garlock fault as indicated by its north-
east trend parallel to it, southeast dip toward it, and by
evidence of left lateral movements on both. The bending
of the great San Andreas shear zone at both points
where it is, or may be, intersected by the Garlock and
White Wolf faults implies that both these northeast-
trending faults are deep seated zones of weakness along
which tile rigid Sierra Xevada-Great Valley block is
being pushed southwestward relative to the Mojave Des-
ert block.

3. KERN CANYON LINEAMENT

By Robert W. Webb •

Introduction. The recent earthquakes in the Tehach-
api area of the southern Sierra Nevada have refoeused
attention of }i-eolog:ists on this eritieal area of California
strueture. The reeent summary of the Arvin-Teehaehapi
earthquake (California Division of Mines, 1952), calls
attention to what may be a structural pattern in a
series of faults (Jenkins, 1938; Nugent, 1942) whose
freolofjieal relationships have never been established. The
faults in question are known as the White Wolf, Breck-
enridge Mountain, Ilavilah Valley, Hot Springs, and
Kern Canyon faults. The regional topographic pattern of
these faults and the inter-segments between them will be
referred to as the "Kern Canyon lineament." It seems
pertinent to examine what is known currently about the
structural pattern and to sviggest a possible interpreta-
tion for the apparent pattern.

Geography of the Faults in the Lineantent. The dis-
connected fault zones and inter-segments that appear to
compose a structural lineament extend from the Tejon
Hills in the southern San Joaquin Valley, northeastward
and northward for more than 100 miles, beyond the
headwaters of the Kern River. The faults have been
studied, and evidence (Hoots, 1930, pp. 301-319; Law-
son, 1906; 1904, pp. 291-376; Webb, 1936) for them pre-
sented. Between these are apparently nnfaulted seg-
ments, none of which has been mapped in sufficient
detail to prove positive connection, at least in the present-
day structural pattern; other inter-segments are un-
mapped.

Historical Background. The first recognition of an
important fault in the Kern River Canyon was by
Lawson (1904), in the first of a series of three papers,
discussing faulting in the upper Kern Basin. In a second
paper (1906) he presents his observations made in the
middle Kern Basin, the Havilah Valley, and Walker
Basin, which suggested to him the apparent importance
of faulting. A connection between the northern (Kern
Canyon) faults and those in the Havilah and Walker
regions was tentatively postulated. In his third paper
on the Tehachapi Range (Lawson, 1906a) he recognized
the important Tehachapi (White Wolf) fault, and raised
the possibility of a connection between the earlier de-
scribed faults and the White Wolf, although no con-
nection between any of these faults was seriously implied,
since he did not undertake geologic mapping. In 1922,
the publication of a structural map of California (Seis-
mological Society of America, 1922) showed the White
Wolf fault, and that part of the Kern Canyon fault
from the mouth of Golden Trout Creek nearly to Fair-
view, as "dead fault, well located;" the Breckenridge
fault is symbolized as "probable fault, character and
location uncertain." Additional geologic studies were
not published until 1928, when faulting was mentioned
incidental to other geologic problems (Hake, 1928; Mil-
ler, 1931). The White Wolf fault was mapped in 1930
(Hoots, 1930), and the Kern Canyon fault studied in
1936 (Webb, 1936). Interest in damsites along the Kern

RELATIONS OF FAULTS IN

KERN CANYON

LINEAMENT

(

FouH

• Professor of Geology, University of California. Santa Barbara.
Manuscript received for publication December, 1952.

FlQDKE 1.

River was revived, and several reports appeared * (Mar-
liave, 1938; Treasher, 1949, 1949a). Significant informa-
tion on faulting in Ilot Springs Valley, at the site of
the new Isabella Dam near the junction of Kern, and
South Fork of Kern River will appear with the full
publication of these studies. A geologic map of the Kern-
ville <iuadrangle was published in 1940 (Miller and

1 Holdredse, Clair, Personal communications, July 19, 1949, and
Oct. 7, 1949.

(35)

Earthquakes in Kern County, 1952

[Bull. 171

Webb), and included the southern part of the Kern
Canyon fault and some observations on the Hot Sprinfis
fault ; the influence of faulting in the mines in the vicin-
ity of Kernville was discussed in a paper on the Big
Blue mine (Prout, 1940) ; and the northern section
of the Kern Canyon fault was mapped in 1946 (Webb,
1946). It is evident that no published work to date
ju.stifies the assertion of a single fault ; nor of the con-
nection of the separate faults into a master structure.

The map of California accompanying the text on the
Arvin-Tehachapi earthquake (California Division of
Mines, 1952, p. 2) is the first known to the writer since
the 1922 map cited above that clearly shows the White
Wolf fault as a separate and distinct structure, based
on geologic mapping by Dibblee (Dibblee and Chester-
man, 1953) in one of the critical inter-segments.

Interpretation of the Lineament. Although the rela-
tionship between the WTiite Wolf, Breekenridge, Havi-
lah Valley, Hot Springs Valley, and Kern Canyon
faults is imperfectly known, the topographic patterns
suggest a regional plan of the faults. The relation to the
curving southern end of the Sierra, where the Sierra
Nevada fault system and the Garlock fault system
merge, seems significant. Also, if such superficial evi-
dence as slickensides and epidotized joints is accepted,
faulting can be demonstrated in segments between rec-
ognized faults. Reconnaissance in intersegments shows
that the rocks involved are almost exclusively of the
massive plutonic type, whereas faults are reflected in
areas where the basement complex includes many pen-
dants and residual areas of metamorphic rocks.

Faults arranged in a lineament, with evidence of
ancient or recent movements on some faults and little or

no evidence of faulting in rocks of inter-segments might
be explained by a geologically ancient continuity of the
faults, with no necessary current relation of faults of a
former continuous structure. An explanation of an ap-
parent regional pattern might be found if the nature of
fault decline with depth were understood.

The roots of a major fault of regional proportions
originally continuous in a rock cover now stripped from
an uplifted land mass, might be evidenced by residual
segments, parts only of the original fault. The struc-
tural effects of fault movement, such as measurable dis-
placement and drag, slickensides, gouge, and breeciation,
effective to differing depths and in differing degrees as
the original fault descended into changing rock types of
the block would thus be discontinuous.

The Sierra Nevada might be such a block, thus faulted
and denuded.

The assumption so often repeated in map and text
of a major continuous fault, extending from the head-
waters of the Kern River on the north to the San Joa-
quin Valley on the south through more than 100 miles
will not easily be dispelled. Though the possibility of a
direct present relationship between separate faults
known in this regional lineament must be admitted, the
concept of a single structure important in today's geol-
ogy should be abandoned. The suggestion that the faults
known today may be remnants of an ancient, and origi-
nally continuous, fault, developed in the cover rocks of
the ancient Sierra Nevada, is advanced as a plausible
explanation of the topographic lineament. Understand-
ing of this lineament, like so many of the faults of re-
gional geologic maps, rests, like all other geologic prob-
lems, on completion of detailed geologic mapping.

4. NATURE OF MOVEMENTS ON ACTIVE FAULTS IN SOUTHERN CALIFORNIA

By Mason L. Hill •

Ahufiact. The i)rinciiial fault typos of snutliprn California are
right lateral, left lateral, reverse, ami thru.>it faults. Some of the
iiortlnvesttreiKlinj; right lateral ami east-mirtlieast-tremlins; left
lateral faults are known to have strike-slip movement. Some of
the reverse and thrust faults are likewise known to have mainl.v
(lip-slip movement. Many of the faults are proliahly characterized
hy ol)li(pie-slip movements where both the dip-slip and strike-slip
components are relatively substantial.

Most, if not all, faults in southern California are potentially
active. Movements on the large lateral faults, which have sub-
stantial .strike-slip components of movement, are possibly respon-
sible for much of the strong seismic activity in this region. Many
of the rever.se and thrust faults of outcrop, especially those asso-
ciated with folded sediments, probably die out before reaching the
focal depths of important southern California earthquakes.

The active White Wolf fault, being parallel to the left lateral-
slip Garloek fault and conjugate to the right lateral-slip San
Andreas fault and being steep and deep, is probably characterized
by a substantial left lateral component of moyement.

Introduction. The Arvin-Tehaehapi earthquake of
July 21, 1952, resulted from a movenient of uncertain
nature on the supposedly inactive White Wolf fault.
Thus we were shockintrly challenged with important
tectonic problems. In this case : what was the nature of
the movement and why did it cause a major earthquake?
In the general case : which other faults in southern Cali-
fornia could develop similarly surprising seismicity?

The principal objectives of this discussion are to indi-
cate that the true nature of movements on, and the cur-
rent activity of, faults are difficult to determine, and
that lateral faults, with substantial strike-slip compo-
nents of displacement, are possibly the most important
breeders of major earthquakes in this region.

The writer acknowledges the use of considerable Rich-
field Oil Corporation field mapping, mainly by T. W.
Dibblee, Jr., and the advice of Hugo Beuioff, Marie
Clark and Rollin Eckis, who have read the manuscript.

Fault Terminology and Classification. A fault is a
fracture in the earth along which movement has oc-
curred. Faults are more or less planar with strikes in
any direction and dips from horizontal to vertical. The
prime criterion for recognition of a fault is offset (sepa-
ration) of geologic features. Although the orientation is
rarely ascertained, relative displacement is described
as dip-slip, strike-slip, or oblique-slip with reference to
the attitude of the fault plane (zone). For faults other
than vertical, the block above the plane is called the
hanging wall and below the plane, the foot wall. Faults
are caused by rock-rupturing stresses but the nature
(tensional, comprcssional, etc.) and origin (gravity, con-
traction, etc.) of these stresses is very rarely known.

A purel}' geometric classification (more practical than
genetic or slip classifications) based on apparent relative
displacement (separation), is a follows (Hill, 1947) :

Normal : In vertical section, hanging wall is relatively
and apparently down — including vertical faults.

Reverse : In vertical section, hanging wall is relatively
and apparently up — restricted to faults that dip
more than 45°.

' Geolosist, Richfield Oil Corporation, Los Angeles, California. Man-
uscript submitted for publication August, 1953. Published by
permission of the Richfield Oil Corporation,

Right lateral : In horizontal section, side opposite ob-
server is relatively and apparently to the right.

Left lateral : In horizontal section, side opposite ob-
server is relatively and apparently to the left.

Thrust : Any fault which dips less than 45° and evi-
dences dip-separation and horizontal shortening.

Right (or left) lateral normal (or reverse or thrust) :
These names for the six combination types are rec-
ommended when both strike and dip separations are
known.

Because of the scope of this paper, which deals with
the relative sense of movements on active faults, and
because there is evidence of the nature of the movements
on some of the discussed faults, it might appear ad-
visable to use here a relative movement classification
(dip-slip, strike-slip and oblique-slip faults). However,
since in the usual ease there is no definite evidence of
the nature of movement, and since in some of the cases
discussed the evidence is not conclusive, the geometric
classification (based on separation) is used, with the
addition of the relative movement (slip) terms where
possible.

Faults in Southern California. The major fault of
the region is the San Andreas. This is, however, only
one of a set of northwest-trending right lateral-slip
faults present. Other principal sets are east-northeast-
trending left lateral-slip faults (e. g., Garloek fault),
and east-west-trending reverse (e. g., the Oak Ridge
fault), and thrust (e. g., the Santa Susana fault) faults
of dip-slip movement. This grouping is believed to be
significant with the right and left lateral faults of east-
west relief ^ resulting from north-south shortening as
the primary strain system - of the region. The east-west
reverse and thrust faults of upward relief result from
the same north-south shortening. Other classes and
trends of faults occur, but in southern California
they do not appear to be major or primary ^ structures.

Obviously the above grouping is tentative because data
on significant fault characteristics are woefully incom-
plete, especially orientations of displacement and even
locations, extents, and dips of the faults themselves. Also
inadequate are data on ages and cumulative displace-
ments on important faults. Other complicating situa-
tions comprise the determination of the relative impor-
tance of dip-slip and strike-slip components on faults of
oblique-slip movement, and the gradation of a fault,
along either strike or dip, from one geometric type to
another.

A noteworthy aspect of some of the faults in southern
California is their transection of geologic provinces. For
example, the San Andreas fault cuts through the Coast
Ranges, Transverse Ranges and the Colorado Desert
without being importantly influenced by diverse rock
types and structures. It appears significant that, al-
though most of the faults are confined to separate prov-

' Orientation of maximum relative elongation of the deformed unit.
- The strain system is the unit of deformation which is caused by a

single but less readily determined, stress system.
3 Primary structures are those faults which are considered to be

caused directly by the regional stress system.

(37 )

Earthquakes in Kern County, 1952

[Bull. 171

a
o
a

o.
a

Part I

Geology

Earthquakes in Kern County, 1952

[Bull. 171

inces, only the laterals extend through more than one
geologic province.

Nature of Movements on the Faults. The orientation
and especially the cumulative amount of movement are
obviously the most critical geologic aspects of faulting
but, unfortunately, the most difficult to determine. Direct
evidence of orientation is occasionally manifest at the
time of earthquakes. Examples are the right lateral
movements of 21 feet (1906) and 10 feet (1940) on the
San Andreas fault. Recent movements are also sometimes
shown by topographic anomalies, such as scarps and
offset drainage lines. Contrarily the sense of movement
along the White Wolf fault was not definitely revealed
by surface displacements at the time of the Arvin-
Tehachapi earthquake nor is the movement clearly re-
vealed by topographic features.

Geologic evidences of orientation are occasionally
shown by fault zone features such as striations, fracture
cleavage, etc., and adjacent drag folds, feather joints,
etc. Other indirect evidences of the nature and amount
of movement on faults comprise offsets of rock or struc-
tural units. These offsets ordinarily show oidy the appar-
ent relative displacement and they are usually described
in terms of the vertical component (throw). In the case
of some lateral faulting, however, indications of sense
and amount of strike-slip displacement may be shown by
sidewise offsets of the following: basement rock facies,
sedimentary facies, stratigraphic thicknesses or se-
quences, unconformities, faunal facies, deposits from
source areas, and offsets of structures.

Recently these criteria have been used to indicate
20 ± miles of right lateral movement on the San Ga-
briel fault during late Miocene-early Pliocene time
(Crowell, 1952a) and possibly several hundred miles of
right lateral movement on the San Andreas fault since
Jurassic time (Hill and Dibblee, 1953).

As described elsewhere in this bulletin, tlie minor (2
feet, plus or minus) displacements along and adjacent
to the trace of the White Wolf fault at the time of the
earthquake indicated, although not clear and consistent,
thrust and left lateral movements. However, since the
epicenter plot for the 10-mile deep focus is near the
southwestern projection of the surface trace, it is ob-
vious that the fault zone is steep. Therefore thrusting
is probably only a surface manifestation, possibly devel-
oped as a creep effect by the sharp topographic relief
between Bear Mountain and the floor of the San Joaquin
Valley.

Some topographic evidences of late activity on the
White Wolf fault are present, although apparently no
worker had suspected it as a seismic threat. The obvious
topographic feature along the fault is the nortliwest-
facing Bear Mountain scarp. Also present are saddles
and sag ponds (connnon on lateral-slip faults) and sharp
left lateral bends of Sycamore and Little Sycamore Can-
yons at their mouths.

The prominent topographic and geologic offset is the
relatively upward throw of thousands of feet of the
southeast (hanging wall) block. This throw, however,
may be a relatively surfieial component of mainly strike-
slip movement at depth and/or the result of the juxta-
position of topograpliically high and low blocks by such

sidewise movement. In fact, the eastern edge of lower
Miocene marine sands appears to be offset several miles
in a left lateral sense by the White Wolf fault (see fig.
2), Eocene sediments may have been shifted even fur-
ther and the upper Miocene Santa Margarita sand can
easily be offset by movement which comprises a greater
lateral tlian vertical component. Because of the above,
and since this deep and steep fault is essentially parallel
to the left lateral-slip Garlock fault, it is here tentatively
classified as a left lateral reverse (obliciue-slip) fault
with the cunuilative amounts of strike-slip and dip-slip
components as yet undetermined. Furthermore, since the
Wliite Wolf fault is nearly perpendicular to the right
lateral-slip San Andreas, to which it may extend under
the south-dipping thrust faults of the San Emigdio
Mountains, it is probably conjugate to it.

Activity of the Faults. Now that we know that the
White Wolf fault is active, whereas it had generally
been placed in the inactive category, a new inquiry into
the locations of potentially active faults seems appro-
priate. Faults in this region commonly show physio-
graphic evidences of geologically late movements by
scarps, trenches, ridges, offset drainage lines, etc. Fur-
thermore, since southern California is tectonically active,
nearly all faults are subject to a possible renewal of
movement and still other faults could be developed. It
appears, however, that movement on certain classes of
faults may be responsible for most of the sizeable earth-
quakes in this region.

Major lateral-slip faults are characterized by great
length and steepness and probably extend to depths of
at least 10 miles. The reverse and thrust (dip-slip)
faults, the other principal types of the region, are rela-
tively restricted in length, are most common in areas
of thick sedimentary sections and may ordinarily ex-
tend only to relatively shallow depths. Therefore, in this
region lateral-slip faults, of which the White Wolf is
possibly a member, are perhaps the best candidates for
the generation of major earthquakes (above 6 on the
Richter magnitude scale and instigated at a depth of
approximately 10 miles). However, even if these are the
likely earthciuake faults, there are so many of them
(without including the probable many others which do
not reach the surface) that there is no tangible reason
for suspecting earthquakes from movement on any par-
ticular fault, or in any specific area, in southern Cali-
fornia.

Conclii.'iions. Tentative answers to the questions of
the first paragraph of this discussion are: (1) a com-
ponent of left lateral movement probably occurred on
the White Wolf fault, and being a deep fault, such
movement occurred at the correct depth and involved
sufficient energy to cause a major earthquake; and (2)
nearly all of the major right and left lateral-slip faults
of the region are seismic threats.

A further conclusion is that the determination of
sense and cumulative displacement on lateral-slip faults,
combined with seismic and geodetic data, are likely to
reveal imjiortant facts and concepts regarding the geo-
logic history and present tectonic status of the region.

5. GEOLOGICAL EFFECTS OF THE ARVIN-TEHACHAPI EARTHQUAKE

By John P. Buwalda • and I'ierrk St. Amand '

ABSTRACT

The Arvin-Tphaoliiipi ciirUuiuakf of July 21, 19,T2, oripinatod
on the White Wolf fault. This fault runs from west of Wheeler
Ridge to the vicinity of Harper Peak. The strike is roughly
N 50° E; the length is nt least 32 miles and it appears to he a
steep reverse fault or a thrust. The overall movement seems to he
oblique slip, up dip, with a left lateral component of motion. The
vertical offset is greater than lO.tHK) feet.

The geologic effects included landslides, rock falls, changes in
ground water and stream How, lurches and fault trace develop-
ment. A series of ground ruptures extended intermittently along
the length of the fault, except across the alluvium of the San
Joaquin Valley, where lurching was developed. At the foot of
Bear Mountain the traces were compressional, indicating thrust-
ing of the southeastern lilock over the valley, coupled with a
small component of right lateral movement. Near the White Wolf
Ranch a left lateral tear fault crossed the upper and lower blocks
of the White Wolf fault. To the northeast of this the fresh
displacements on the White Wolf fault were primarily left lateral
and tensional. There are, in places, minor excejitions to the
general displacements and nearly all the traces are complicated
by landsliding.

INTRODUCTION
The Arvin-Tehachapi disturbance was the strongest
in California since the Han Francisco earthquake of
1906 and in southern California since the Fort Tejon
shock of 1857. Because some of the stronger earthquakes
in California and Nevada dvirinp; the past century were
accompanied by surface displacements and other geo-
logic and physiographic changes along the faults on
which the shocks originated, it was hoped that similar
features would be found along the Wliite Wolf fault.
This expectation was realized only in part, and the
features developed were quite different from those pro-
duced in the San Francisco earthquake of 1906, Im-
perial Valley in 1940, Owens Valley in 1872, and Pleas-
ant Valley, Nevada, in 1915.

Immediately after the earthquake, instrumental
parties, under instructions from Dr. Beno Gutenberg,
Director, and Dr. C. F. Richter, Seismologist, of the
Seismological Laboratory of the California Institute of
Technology, located mobile seismographic units set dif-
ferent and changing points in the earthquake area to
record aftershocks with a view to securing evidence on
exact location, extent, and mechanism of the faulting,
depth of foci, and other problems. The authors, at the
same time, started an intensive, systematic field investi-
gation of all the geologic and physiographic changes
that occurred at the time of the earthquake along the
causative fault. This work continued intermittently for
2 months, and was concentrated along the fault zone.
Rupture and other phenomena were abundant for sev-
eral miles on either side of the fault and some attention
was given to them, but field work was terminated when
it was realized that all of the myriads of surface evi-
dences of ground disturbance could not be studied — the
effort had reached the stage of decreasing returns for
time invested.

A brief field examination was also made along the
trace of the northwest-trending Kern River fault after
a rather strong shock, iflagnitude about 6.5, apparently

• Division of Geological Sciences, California Institute of Technology.
Ed. note : This paper was set in type after Dr. Buwalda's death
in August, 1954.

occurred on it on July 29, 1952. No evidence of surface
fault displacement was found. All the phenomena re-
corded in this paper are believed to relate to the Arvin-
Tehachapi earth(|uake of July 21, 1952 and possiblv to
aftershocks centered near the White Wolf faidt zone.

Hundreds of ruptures cut the alluvium on the floor of
the entire southern end of the San Joaquin Valley at
least as far north as points west of Pixley, which is
some 45 miles north-northwest of Bakersfield. The au-
thors did not attempt to map these, but other geologists
-have made careful studies of them in some areas (Warne,
Part I, Contribution 6, this bulletin).

In the field study the ground ruptures along the White
Wolf fault were traced and mapped carefully from
Tejon Hills to Centennial Ridge, 4 miles northeast of
Caliente.

Similar features were examined on Wheeler Ridge,
toward the south end of the fault, and on Harper Peak,
which is 9.5 miles northeast of Caliente and possibly
near a northeastern extension of the fault. Attention
was also given to a number of localities showing other
unusually interesting ground fractures, among them the
south end of Walker Basin, the west side of Brecken-
ridge Mountain, the higher parts of Bear Mountain, and
a short section on the Garlock fault where the Oak Creek
Pass road crosses it.

Two preliminary papers relating to the 1952 Kern
County earthquakes were published (Benioff, et al.,
1952; Buwalda and St. Amand, 1952).

LOCATION AND EXTENT OF WHITE
WOLF FAULT

The Arvin-Tehachapi earthquake originated on the
White Wolf fault, at the south end of the San Joaquin
Valley. This crustal fracture is known to extend from
Wheeler Ridge, in the middle of the south end of the
Valley, on a course N. 50° E., to cross the eastern margin
of the valley to or beyond a point on Caliente Creek 1.0
mile northeast of Caliente. Its known length is therefore
about 32 miles. The White Wolf fault was first shown
on a map by A. C. Lawson (1906a, facing p. 432) but
not mentioned or named by him. It was named and de-
scribed briefly by H. W. Hoots (1930) and its position
was indicated on his geologic map for about 5 miles
northeastward from Comanche Point. Previous to the
earthquake the surface geologic evidence for the exist-
ence, location, and attitude of this fault was rather gen-
eral. From Comanche Point northeastward, in the
Tejon Hills, the inferred fault is roughly a Ijoundary
between valley alluvium and the Tertiary formations.
Farther northeast it is either a boundary between valley
alluvium and the old crystalline rocks of Bear Mountain
or lies with old crystalline rocks on both sides. The one
possible exception is a fault contact between Tertiary
strata to the northwest and the old crystalline rocks on
Caliente Creek 1.0 mile northeast of Caliente, but this
point maj' be northwest of the fault. From the Tejon
Hills northeastward the zone of the fault trace has
nearly everywhere suffered enormous and widespread
landsliding from Bear Mountain scarp. The usual types
of geologic evidence for tracing a fault between Tejon

(41 )

Earthquakes in Kern County, 1952

[Bull. 171

Hills and Caliente Creek, some 16 miles, are therefore
missing or have been obscured. The existence and gen-
eral location of the White Wolf fault was inferred
originally from the bold scarp forming the northwest
face of Bear Mountain, rising 4,000-5,000 feet above the
floor of the San Joaiiuin Valley, and the depression along
its base in which the White Wolf Ranch is located. Since
the base of the bold northwest face of Bear Mountain is
not a straight regular line, such as marks many steep
fault scarps, mainly because of landsliding, and there is
little contrast in rock types in the scarp and in the foot-
hills, the exact location and course of the fault trace has
not been, and is not now, determinable so far as the
writers are aware. It has been located only roughly, and
northeast of Comanche Point entirely or almost en-
tirely on physiographic evidence.

Geologists have long suspected that the White Wolf
fault is a southwest-trending terminal segment of the
important Kern Canyon fault or fault zone which is
followed for over 70 miles by the south-flowing main
fork of Kern River from north of Mount Whitney to
Kernville and which then, with a more westerly branch
(Breckenridge fault) as shown on Dibblee's map, forms
the high scarp bounding Walker Basin on the west. It
is not yet certain that this is not the true relation but it
appears from the distribution of aftershocks and of
ground ruptures that the White Wolf fault probably
continues northeastward beyond the point where the
Kern River or Breckenridge fault projected southward
would meet it. Although it has not been possible to trace
the Breckenridge fault to the White Wolf it is improb-
able that such a long and important fault zone would
terminate only a few miles from an intersection with an-
other important fault.

While it has not been possible to trace the White Wolf
fault northeastward by ordinary geologic or physio-
graphic evidence beyond Tehachapi Creek, it may be
more than a coincidence that a quite strong earthquake,
magnitude 6^, occurred on 15 March, 1946 north of
Walker Pass, about 44 miles to the northeast and on or
close to its projection. No fault is known at the epicen-
ter of the Walker Pass earthquake.

Reverting now to the southwest 12 miles of the White
Wolf fault, from Comanche Point to Wheeler Ridge, it
is not indicated by surface evidence. This deeply allu-
viated plain at the south end of the San Joaquin Valley
showed no scarp or warped surface, so far as known, to
mark the course of this important fracture, either pre-
vious or subsecjuent to the earthquake. But geophysical
studies demonstrate the existence and general location of
the fault very clearly. Its position as plotted on the map
(plate 2) is based on seismic reflection data and was
kindly furnished by Mr. Rollin Eckis, Chief Geologist,
and Dr. Mason Hill, Senior Geologist, of the Richfield
Oil Corporation of Los Angeles. The line shown is not
the surface outcrop of the fault but its approximate
trace, within the limits of geophysical accuracy of loca-
tion, on the surface of the granitic basement. The fault
in all probability dips southeastward and the line shown
is therefore presumably the northwestern overhanging
edge of the granitic block southeast of the fault. Prom
the surface near the mouth of Sycamore Canyon, north-
east of Comanche Point, the trace descends to an eleva-
tion of nearly 8,000 feet below sea level at a point about

3.5 miles southwest of the point. The bedrock trace has
apparently veered to a position roughly three-quarters
of a mile southeast of the southwestward projection of
the nearly straight section of the fault as inferentially
traced from Sycamore Canyon northeastward past White
Wolf Ranch to the neighborliood of the railroad ; since
it dips south this is expectable. In the next 6 or 7 miles
southwestward the bedrock trace, in approaching the
northeast face of Wheeler Ridge, rises from about — 8,000
feet to about — 3,000 feet. While the geophysical data here
are less exact the trace does not appear to return toward
the southwestward projection of the surface trace of
the fault, but continues sub-parallel to it. If a fact, this
is explicable either bj' a change to a slightly more south-
erly strike, as it proceeds southwestward, or to a slight
flattening of the dip of the fault. The course of the
surface trace of the fault as inferred northeast of Syca-
more Canj'on, projected southwestward, would pass
slightly north of the highest point on Wheeler Ridge,
and the projected bedrock or seismic trace would pass
a bit south of it. It is interesting that a zone of surface
ruptures occurs on Wheeler Ridge on the southwestward
projection of the bedrock trace ; the zone also has about
the same trend as the strike of the fault. Presumably the
steeper White Wolf fault passes southwestward under
the lower, south-dipping Wheeler Ridge overthrust and
the Pleito thrust south of it, both of which trend more
nearly east-west than the White Wolf. At any rate, in
spite of the fact that the epicenters of the main shock
and of the one foreshock were somewhat south of the
highest part of Wheeler Ridge, no evidence of the White
Wolf fault southwest of the ridge was found. Prom the
point where the bedrock trace of the White Wolf fault
passes under the northeast face of Wheeler Ridge it is
about 17 miles measured along its southwestern projec-
tion to the San Andreas fault. One might well suspect
that a fault with as large displacement as the White
Wolf would continue at depth to an intersection with
the San Andreas, which it would meet at an angle of
about 60°, but this can apparently only be speculation
at present.

It is interesting that the White Wolf fault trends
roughly at right angles to the Kern River, Bena, Tejon,
and other northwest-striking faults of the eastern part
of the southern end of the San Joaquin Valley.

Although the magnificent northwest face of Bear
Mountain obviously resulted from relatively recent ver-
tical fault movement at its base, the White AVolf fault
had not generally been considered one of the State's
active fractures, and an expected source of earthquakes,
by geologists in the past. No scarplets due to late dis-
placement along the base of the scarp, such as occur
at numerous points along the south base of the San
Gabriel Mountains, had been noted so far as is known,
and the course of the fault under or through the allu-
vium between Comanche Point and Wheeler Ridge is
not known to be marked bj' such evidences of recent dis-
turbance as scarplets, sagponds, trenches, and drainage
derangement so common along the San Andreas and
other major active faults.

GEOLOGICAL EFFECTS OF THE EARTHQUAKE

When movement on the White Wolf fault occurred on
the morning of July 21, 1952, an interesting expression

Part I]

Geology

of that movement developed aloiifj the fault trace. For
nearly 40 miles, a suceessioii of features raufriiif:' from
lurch cracks to actual fault displacement marked the
position of the fault zone. In many places the features
were obscured and complicated by landsliding and
slumpinfr and in others cross faulting develoj)ed on an
impressive scale. The following account is a detailed
descrijition of the phenomena developed along the sur-
face expression of the "White Wolf fault.

Chronologically, attention was first attracted to the
fault zone in the region of Bealville where damage was
done to the railway tunnels and along the Arvin cutoff
roatl where conspicuous scarplets were developed. This
account presents the observations in a geographic se-
((uence, beginning in the epicentral region and continu-
ing in a northeasterly direction to where the fault zone
dies out in the region of Caliente. The reader may find
the map of the fault trace (plate 2 in pocket) helpful
as references, by number, are made to specific localities
thereon.

Han Kmiffilio Ranch. The shakinj; at the San EmiKili" Ranch,
near the western end of Wheeler Ridge was .severe, oau.sing dam-
age to structures and developing a numlier of gaping furrows 6
inches wide and 200 feet long near the ranch hoiise. The cracks
were sub-parallel to the contour lines and were best developed in
irrigated, filled land. A number of 1- and 2-inoh water pipes l.ving
on flat alluvium were ruptured ; similar pipes on nearh.v hillsides
were not. The fissures were deemed to be lurch cracks caused by
severe shaking rather than by actual fault displacement.

Wheeler Ridge. The most southwesterly ground ruptures that
are probably more or less directly related to the movement on the
White Wolf fault during the Arvin-Tehachapi earthquake, rather
than merely to the lurching which presnmalily produced most or
all of the cracks on the floor of the San Joaquin Valley, lie in a
narrow northeast-southwest zone obliquely across the upper half
of the highest part of the east-west Wheeler Ridge. This ridge is
a more or less isolated feature rising 1000-l."i00 feet above the
flat floor of the south end of the San Joaquin Valley ; it lies im-
mediately west of the main Los Angeles-Iiakersfield highwa.v —
the Ridge Route. The ruptures are of particular interest for sev-
eral reasons. Ground distortions and fractures, more or less di-
rectly related to the fault movement, which are so conspicuous
from Caliente to Tejon Hills, are apparently absent from the 12
miles of flat San Joaquin Valley floor southwest of the Tejon
Hills, but apparently reappear in Wheeler Ridge. They occur
on the part of the Ridge which is on the .southwestward pro.jec-
tion of the fault as plotted from geophysical data, and very few-
ruptures occur on other parts of the Ridge. Their trend is that
of the fault. Their trend projected southwestward pas.ses close to
the instrumental epicenter of the main shock.

As in the Tejon Hills-Caliente section of the fault zone there
are at least three types of ruptures on Wheeler Ridge. The most
numerous are .soi'. cracks; these are often tens of feet long, occa-
sionally one or two hundred feet long, and tend to be parallel to
the contour lines. They are often rather widely open and are
clearl.v due to a thin layer of soil, not over a few feet in thick-
ness in most cases, slipping directly down the bill slope on the
firmer rock surface on which it rested. Some of these cracks devel-
oped in other parts of Wheeler Ridge also. A second type occurs
around the upper end of old or new landslides ; it is more curved,
the horns of the arc pointing down hill. These ruptures are clearly
the result of movement or resumption of movement of landslide
masses, the upper parts pulling away from the stationary ground
above. The landslide ruptures were most numerous near the main
northeast-southwest zone of ruptures ; the local direction of slope
of ground determining the direction of landslide movement and
hence the trend of the ruptures.

The third and most important of the breaks were the long
straight ones which crossed the crest of the hills obliquely on the
projection and trend of the fault. I'nlike the two previous types
they were independent of topography, traversing hills and depres-
sions indifferentl,T. There were three of these cracks. One began
in the deep canyon draining northward next east of the main
group of Standard Oil Co. wells, at a point perhaps one third of
the way from the crest of the ridge down to the north base. This

crack climbs the hillside with a strike of S. 55° W., and cros.ses
a road at a point 2r)<) feet S. 20° K. of KCL well No. 20. At this
point the crack is clearly an (dd fracture or fault dipping 50°
northwest. .\ layer of gouge about a si.vteenfh of an inch thidc oc-
curs on it. The grooves and striations indicate about eipial dip
slip and strikesli]) movement, right lateral in direction. The
crack continues to the crest of Wheeler Ridge and ends in the
second gulch east of the main oiled road leading to the Standard
Oil camp from the south. It is roughly 3 mile in length. Near its
southwestern end it crossed a tight east-west wire fence at an
angle of .S0-4() degrees ; movement on the crack did not break or
slacken the wires but offset the alignment about 4 inches. This
could have resulted from the 1 foot of dip slip on the fracture.
The crack is nearly straight in general plan but quite crooked in
detail and apparently diil not experience much horizontal move-
ment. I'sually the northwest side of the crack bad dropjied down
1 to 4 feet and the oi)eniug between walls was t> to 12 inches
wide. The rupture clearly cut the Pliocene P^tchegoin formation
and was not due to soil slippage or to ordinary landsliding. The
southwest end of the crack is roughly an eighth of a mile south-
east of the superintendent's house.

A second long crack ends northeastward at the same second
gulch, about 1500 feet east of the oiled road, at which the first
crack ends southwestward, and about .300 feet north of its end.
Crossing the main oiled road 100 feet south of the east-west fence
line and cattle guard about I mile south of the camp, it continues
with the .same S. 50° W. strike across the canyon and beyond the
crest of the ne.xt north-south spur which leads .south from the tree
enclosed su])erintendent*s hf)use on the hillto]). It i)asses about 50
feet north of a huge boulder which lies about 650 feet south of the
superintendent's hou.se. This crack shows no horizontal offset ;
the oiled road surface north of it dropped a few inches relatively.
It is also about three-eighths of a mile long.

A third crack crosses the road about 25 feet north of the east-
west fence line and cattle guard. It is short, and e.xtends only
about 200 feet northeastward from the oiled road. The pavement
was dropped a few inches on the south side of it.

The third or short crack overlaps on the second, hut the two
long cracks do not overlaji — they are not en echelon. The second
or southwestern long crack lies northwest of the southwestern
projection of the first one and begins where it terminates. Al-
together they are over half a mile in length and cross the crest
of Wheeler Ridge. The dominant vertical movement on the cracks
is down on the northwest side, as much as 4 feet ; the horizontal
movement is so small as to be rather uncertain in direction, but
it is probably right lateral.

While the cracks do not extend southwest of the spur leading
.south from the superintendent's house, it is an interesting fact
that the Richfield oil pipeline was ruptured during the earthquake
roughly a quarter to three-eighths mile farther southwest and
approximately on the projection of the zone of cracks. The pipe-
line here lies just west of the paved road ascending the south
slope of Wheeler Ridge and leading to the new Richfield wells ;
it is the paved road next west of the one leading northward to
the Standard Oil Co. wells and cros.sed by the zone of cracks. The
break in the Richfield Oil Corporation pipeline was about 100 feet
north of the telejibone box on a pole along the road. Search failed
to reveal any ground ruptures hereabouts or in the territory to
the southwest.

The zone of cracks and the pipeline rupture presumably are
the surface expression of fractures and sharp distortion which
have extended steeply upward through the Wheeler Ridge over-
thrust plate from the trace of the White Wolf fault below it.

Sail Joncjuin Valley. There were many cracks in the San Joa-
quin Valley floor that were distributed over a distance on either
side of the fault. Most of them were lurch cracks and were not
actual fault displacements. Offsets in cotton rows were common
but as often in one direction as in another, and after a few
hundred feet, the displacement on a particular crack sometimes
reversed. At a spot on the Arvin-Wbeeler Ridge road about 3
miles north of the fault the shaking was especially severe. Large
lurch cracks abounded and a reservoir on the west side of the road
was ruined by them.

Comanche Point. There are at least six prominent cracks be-
tween Comanche Point and Little Sycamore Canyon, numbered
from 1 to 6. The crack at point 1 trends N. 40° E. for 500 feet.
The northwestern side was dropped down 15 inches. There are
en echelon cracks a.ssociated with it indicating a possible, hut
small, right lateral offset. The soil here is thin and gopher holes
are surrounded by fragments of Tertiary shales and red sand-

Earthquakes in Kern County, 1952

[Bull. 171

stone. The crack lies 1(X) feet alxive the inaifiiii uf the valley
alluvium.

A :<fK)-foot t'l-.-K'ture iit point 2. trending N. 40' E.. con.sisting of
en echelon crack.s indicative of a small left lateral motion, shows
a consistent upthrow of 15 inches on the southeast side. This
crack is located 100 feet northwest of the base of the hills.

At 3, the crack lies to the southeast of. and overlaps, the one
described at point 2. This is not a continuous break but con-
sists of en echelon cracks indicative of a small left lateral strike
slip displacement and was accompanied b.v an upthrow of 17
inches on the southeast side. This crack lies at the base of the
hills and is partl.v in alluvium and partly in bedrock. The down
dropping appeared as if the alluvium had settled. In spite of the
indicated left lateral motion, there is but little other evidence of
horizontal movement as the irregularities in the individual cracks
fit together. The trace is curved, and along the base of the hills
the rupture developed into a compressive moletrack .30 to 40 feet
wide and appeared to dip towards the hills at angles of 30 to 60
degrees. This crack does not follow the base of the hills faith-
fully, but cuts across the base of promontories and then over
alluvium-filled valleys and again over the bases of the hills. The
trace dies out in the alluvium of the Comanche Creek fan.

The crack at point 4 is a few hundred feet long, down on the
southern side a fraction of a foot and trends N. 30° E. It is in
the alluvium, about three-quarters of a mile from the fault line.

At the base of the hills, at point 5, is a rupture 200 feet long,
downthrown on the southerly side. Its trend is parallel to the base
of the hills, N. 50° E. and there is some indication of compressive
movement.

At 6, to the east of the fault trace, is a rupture that appeared
to be the result of landsliding. It is 1,000 feet long, trends N.
30° E., was downthrown on the southwest 17 inches, and is usually
continuous. In places however, it developed an en echelon habit
with a right lateral pattern. It lies at the base of a face which
is apparently the back of a landslide mass. This crack is of
interest because it was probably of landslide origin, but resembled
those near {he fault trace.

Little Sycamore Canyon. The northwest end of the smooth,
broadly rounded spur southwest of Little Sycamore Canyon
(point 7) is traversed by about two dozen cracks. Most of these
are parallel and sub-parallel to the contours of the hillside. The spur
is a promontory due to landsliding. Most of the cracks are irregular
and crooked, the average length being 200 to 300 feet with several
up to 900 feet in length. The whole series covers a zone a mile
wide and had the appearance of landslide fractures. This area was
not easily accessible and was examined with binoculars from a
distance of three-eighths of a mile.

Trace Along the Foot of Bear Mountain, East of Arvin. Al-
though this portion of the fault trace is the most continuous and
impressive of all, it is of different character in different places
along its length and at times demonstrates a differing offset. It
will be described in detail beginning from the southwest end. This
trace was first noticeable at point S, approximately half a mile
south of Bear Mountain Boulevard. Here the trace lies in the
alluvium of the San .Joaquin Valley. It is distinguished by a series
of pressure ridges from 2 to 10 feet in length with little or no
evidence for other movement. There was no perceptible change in
elevation across the trace. It was difficult to follow in this section
becau.se of grass, cultivation and trampling by cattle. Where the
moletrack crosses north-south fences it demonstrates right lateral
offsets of a traction of a foot. The trend of the trace in this
section varies from N. 22° E. to N. 35° E. Between 8 and 12, the
traces form a gradually curving track which turns more and more
eastward until it becomes almost east-west at the foot of Bear
Mountain. As it nears the mountain it crosses a road not shown
on the map and in so doing develops a compression crack 0 inches
high and 00 feet long which indicates a right lateral offset.

At point !) and southw^est, the trace is a series of pressure ridges
with a small amount of right lateral movement. At point 10, the
trace developed a clear vertical uplift of 3 to 4 feet on the south-
east side, probably indicating uplift of the mountain. Here the'
trace is a single pressure ridge with a few cracks on the southeast
side. The ridge is essentially a buckle, or a broken warp, without
great evidence of shortening. There was some evidence of right
lateral strike slip movement at this point.

In the vicinity of 11, the trace developed into a series of pres-
sure ridges at the base of a low hill. The total shortening over
four of the.se ridges was estimated as 2 feet. The vertical uplift
was of the onler of 2 or 3 feet. The evidence of horizontal move-
ment was not clear, of the order of au inch or so, and indeterminate

as to sense. The low hill at this [loint is of some interest, having
a lobate shape something like a debris tongue ; it forms a mesa-
like platform at the foot of the mountain. On the western side it
shows evidence of having been trimmed back by washes coming
from the face of Bear Mountain. Xear the west end a gulch in
one of these washes has exposed a section aligned northwest-
southeast. Near the base of this 40-foot section is a 5-foot layer
of gray sand intercalated in the fanglomerate which has boulders
as much as 3 feet in diameter. The sand layer is more nearly
horizontal than the present stream bed and may even dip back
toward the mountain, due to rise of the north edge of this terrace-
like feature. The mole track in general follows the north edge of
this feature in this region and it seems as if the feature had been
formed by previous uiilift along the fault, or at least along the
moletrack.

In places there were multiple fractures; these follow along
berm-like i)latforms on the edge of the so-called terrace.

At 12, the trace makes a 50° turn. Southeast of the turn the
pressure ridges are weaker, the cracks stronger and indicative of
right lateral motion. West of the turn the pressure ridges are
stronger and the cracks absent. The vertical off.set in the pressure
ridges amounted to as much as 4 feet. Conjugate sets of tension
cracks and pressure ridges were well developed along this section
of the scarplet, the pieces lying between them forming triangles
of .sod so that the cracks were behind the part that move forward
and the compression ridges before.

The trace follows the contours at the base of the low hill pre-
viously described and is some distance out from the face of the
mountain proper. The ridge is here 10 to 15 feet above the gen-
eral fan surface.

At 13 the trace is single and is situated on a very gentle slope
near the base of the low flat ridge. Strong cracks are superim-
posed on the pressure ridges in such a manner as to indicate right
lateral motion. The vertical offset here is 15 inches over a zone
comprising seven cracks. The strike of the zone is X. 45° W. as
the trace follows along the foot of the hill and enters each reen-
trant. A few hundred yards up the hill from point 14 is a water
tank some 10 feet across and about .SO inches high from which
a great deal of water was flung during the earthquake. At ])oint
14 the trace is subdued and there was very little vertical dis-
l)lacement. Two or three pressure ridges run parallel to the course
of the trace and several 1- to 2-inch cracks indicate a feeble
right lateral offset.

At 15, the trace is well developed with a zone of cracks about
KHJ feet wide, composed of as many as 12 separate pressure ridges.
The shortening of the original hillside at right angles to the pres-
sure ridges must have been several feet. The trace at this point
may have been complicated by creep of a surficial layer of sod
an<l soil which had slipped down slope as indicated by many
plates of the material which had slip surfaces jiarallel to the hill
slope. Oldham reports a similar effect in the Indian Earthquake
Report, page 111. Approaching point 16 from the west, the trace
swings up and back out of a canyon and in doing so encloses a
portion of a fence set across the canyon. The enclosed portion of
the fence was moved southward up the canyon about 2 feet with

FlGURlc 1. Fence offset by thrusting of hill toward valley.
View northeast near point 16.

Part I]

Geology

r('S|)cct to till- other portion, iiulioatiim that thc> iiioinitaiii li.'ul
moved northwestward with respect to tlie valley l>loi'k.

At point Ki. the traee is still mainly a compressive moletrai-k
and is ac(i>niiianied hy a vertical offset of 4 feet in tln> profile of
the hill. A combination of open cracks and pressure ridjjes is
l)resent which indicated some ripht lateral .shearinf; movement.

At point 17, the trace consists of pressure ridKes 4 to 5 feet
in height over a zone 100 feet wide. The larsest ridge is just
aluive the base of the hill and appears to have a dip of 1")° or le.ss to
the southeast and to offset the profile 2 to ;{ feet vertically. It is
impossible to determine the nature of the lateral movement at
this point, or even if it exists, because of the complicated mole-
tracks and the surficial slumijin;; and slidini; which appears to
have taken place. At each canyon hereabouts the soil slid down
the sides of the canyons into the trough and the whole mass had
migrated down hill as well. Between point 10 and point 17, a
branch trace ran up a canyon as shown on the map. There are
several of these features and they show the same characteristics
as the parent moletrack to a large extent but are not as well
developed. Between 16 and 17. the trail has been crumpled by
the moletrack and little platelets of dried soil were pushed up
to form overthrust ridges. A broken J-inch water pipe crossed
the trace near here and the eiids were offset a fraction of a foot
in a right lateral sense.

At point IS the overthrust character of the moletrack was best
developed. Here a pressure ridge composed of a series of soil
plates was found humped up between 0 and 10 feet. The whole
movement must have taken place during or since the earthquake
because the blades of dried grass when ob.served, were standing
at an angle of 25 to 30 degrees to the vertical, away from the
hill side. It was not possible to judge the amount or sense of
lateral movement at this point. .lust to the northeast, there were
tensional cracks developed which indicated a feeble right lateral
movement.

The trace continues, mainly as a pressure ridge, or moletrack,
with no well defined evidence of shear movement, until a left
lateral displacement may be noted at point 19. Here it loses its
primarily compressional characteristics and demonstrates a 2i-foot
left lateral offset of a fence. This is contrary to the displacement of
all the traces to the southwest where the lateral movement was
primarily right lateral in sense and ver.v feeble.

Eastward of 19, the traee continues over the nose of the spur
on which 19 is located and then turns up the next canyon to the
east. The shaking must have been severe, because boulders near
here, both on spurs and in valleys, have been rocked out of their
nests and vigorously jostled. This was especially noticeable be-
tween points 19 and 23.

The trace makes a sharp turn and proceeds up the canyon,
following it faithfully, although changing at times from one side
to the other. The trace here is double part of the way, and as
it climbs the canyon shows increasing signs of becoming tensional
in nature. At point 20, the trace bounds the east side of a swale
perhaps 100 feet across. Here the trace is an open fissure, down
on the ^vest side on both parts, 3 feet on the easterly part and
one foot or less on the westerly part. Broken roots in the trace
have displaced ends which showed that the westerly side moved
south () inches to a foot. A fence stretched across the swale at
this point, and formerly occupying the bottom of it, has been

^'•t^.ir*.!!

Figure 2. Tensional fracturing. View east near point 20

FiGlBE 3.

Stretched fence at point 20 ; tensional trace
in middle ground.

stretched so that the lower ends of the posts in the middle of
the swale were 7 feet in the air. The posts were formerly buried
for about 2 feet of this length and this makes the uplift of the
bottom of the fence posts at least 9 feet.

Beyond this point, the trace turns south and begins to double
back as if it were ringing a huge landslide mass. The trace dies
out at the point shown on the map, showing tensional qualities and
having stretched another fence at 21, going over a nose so tightly
that it lies upon the ground. Other ruptures at about the 3500
foot level were mostly tensional in nature.

In a number of localities, noted on the map, tensional features
may be seen on the tops of ridges above the moletrack which winds
along the base of hills.

A search near 19 failed to produce any evidence that there
was a moletrack, or fault trace of any sort, going northeasterly
to connect with those farther along the scarp in the vicinity of
the AVhite Wolf Ranch. There was landsliding in this area, one
or two small springs were developed in canyons, and some boulders
rocked out of their nests ; near the highway, the fills and cuts
of which showed considerable mass movement, a large boulder
had been rolled down hill and lay partially blocking a small canyon.

Another trace begins at Point 22 and winds along the base of
a low scarp, up the White Wolf grade, rounding the edge of the
little valle.v in which the White Wolf Ranch is located and climbs
gently toward the hills. This trace is primarily a compressive
moletrack with some indication of right lateral strike .slip move-
ment. The trace goes over several noses and the compression is
enhanced where it climbs their western sides. Near its north-
easterly end, the trace begins to climb and soon dies out on the
steep slopes of Bear Mountain. It could not be followed northeast
of point 23.

Xorth-South Fault Xear White Wolf Ranch. The next trace
is of considerable interest and trends at a large angle to the others.
It is a cross fault with a generalized trend of N. 10°E. The break
is continuous and was easily followed for 3i miles. The predomi-
nant displacement was left lateral and varies from imperceptible
amounts to several feet. The trace could first be discerned at point
24 near the 3600 foot contour on Bear Mountain. The whole
mountain face in the vicinity of the fault is shattered : locally,
sliding and mass movement obscure the trace. The whole mass of
the mountain appears to be crushed and nowhere are any extensive
bodies of solid rock visible. Above the end of the trace, the ground
at about the 4000 foot contour, at an excellent spring, suffered
severe lurching and sliding. There are perhaps 25 tension cracks
between the end of the trace and the spring. Pipes running down
hill from the spring were stretched, but not broken, and many
of them have been pulled so that they are no longer in contact
with the ground. Near 25 the trace offset fences and wheel ruts
in a left lateral manner.

At point 25. the trace bifurcates and the two branches con-
tinue down the mountain to rejoin at point 26. At the 2800 foot
level, the westernmost branch is a crack which is up on the west
from 1 to 4 feet. The crack is often a foot to a foot and a half
wide and en echelon cracks indicate left lateral movement. Many
of the trees here have had branches broken off and one live oak

Earthquakes in Kern County, 1952

[Bull. 171

■•••■^i> -r

Figure 4. Looking north at left lateral offset of fence by
north-striking cross fault at point 27.

tree with a defective l.^-inch trunk was snapped off at the base.
Boulders were rotated by slippage on the moletrack. The material
between the two branches appeared to have slid down hill. The
trace itself seems to be related to minor topographic features,
such as small swales, cols and large terraces.

The feature persists as a strong left lateral moletrack to the
north of 26 and at a point near 27 it leaves the hill and enters
the alluvium of the little valley containing the White Wolf Ranch.
At point 27 it offset a north-south fence in a left lateral direction
some 10 inches, indicating a considerable displacement, even
through some depth of alluvium. The trace was easily followed
northward as a zone of en echelon fractures, and where it crossed
the Arvin road it developed an excellent set of en echelon frac-
tures in the asphalt. The total opening in the cracks, measured
along the center line of the road the day following the earthquake
was 1.1 feet ; the white line was offset 3 inches. Later the road
was patched and on 11 August it was noted that the same set
of cracks had again opened, indicating that movement on this
fault continued for some time after the main sho<k. The fence
wires on both sides of the road were stretched inordinately tight
and the barbs were dragged across the fence posts, scratching
them deeply and even pulling out some of the staples.

North of 28 the trace closely follows a small stream course as
it crosses the ne.xt field. It appears that the stream course had
been determined by previous movement on this fault, as its
course was very straight and ran across the regular drainage
pattern. A former crack had been utilized by the rill in the past
and was reopened by the eartbf|uake. Where the trace crosses
highway 466 at 2!), the pavement had been patched and no measure
of offset could be taken. The trace crossed 466 at a culvert, still
following the water course. North of this point the trace was
still distinct and could be followed as a set of en echelon left

FllU UK

l!n fchfli'ii cracks in Arvin road at point 28,
indicating left lateral movement.

lateral cracks which continued to point .30. located in a col between
two hills.

This cross fault is an important structure and appears to exist
in both the upper and lower blocks of the White Wolf system.
Perhaps it is primarily a lower block feature and was extended
into the upper block l)y frictional forces but it may actually cut
the White Wolf fault in a primary sense, dividing both blocks
into two parts.

Between the White Wolf Ranch and Rogers Ranch. The trace
at H\ appeared to be greatly complicated liy landsliding, the sliding
utilizing the trace as an upper boundary and modifying it. Here two
ruiitures run around the nose of the spur above the edge of
valley alluvium. The ruptures are sub-parallel to the contours.
The upper one is an open vertical fissure, the lower one an over-
thrust, flat fracture in the soil. The two ruptures gradually die
out toward the canyons on either side. There is an indication of
left lateral .shear along the upper, or tensional feature, the north
edge of which dropped down a fraction of a foot.

Farther up the same slope near 32, the whole slope is shattered
and there are a number of features, primarily tensional, upon
which no left or right lateral movement was noted. There may
have been some lateral movement, but the features appear to be
the result of landsliding.

A trace heading northeastward was noted near the end of this
fracture ; it continues to the next mole track to the north, at 33,
and beyond. There is some evidence that this crack, which also
looks like a landslide feature, but which shows left lateral and
extensional movement near 33, extends farther to the north, as
the road in the col 34 lietween the sawmill and the White Wolf
Ranch was fractured, as was the Arvin road. 35. one (luarter mile
west of tlie junction with highway 466, and 466 itself was frac-
tured at a point N. .~>° W. of this point. The indication is that
a zone of fracturing of considerably less, and indeterminate, offset
crosses the roads and perhaps joins with the north-trending cross
fault that passes near the White Wolf Ranch. This trace is dotted
on the map to indicate that it is not definitely conuectable through-
out the whole distance.

The fracture at 36 is jirimarily a compressional feature with
some evidence of left lateral displacement. Pipes crossing the trace
are bent and show a general shortening of the area. The edges of
the cracks, and occasional en echelon cracks and pressure ridges
indicate left lateral movement. The trace is confused near point 33
and terminated in the vicinity of the northeast-trending fracture at
that point. To the east, the trace reached a maximum displacement
and development in the center of the ridge it transected.

Farther up the mountain face at 37, is another trace, removed
a considerable distance from the general zone of fracturing. The
north side of this rupture was uplifted 6 inches to a foot. There
is a spring at both ends and in the middle of it. This extensional
trace was first noted liy C. R. Allen on 13 September and may not
have been made by the main shock, as it was not noticed earlier,
when the area was examined, although it is quite possible that it
was overlooked. This feature disappears in the detritus in the
canyon at both ends.

From 38 eastward the rupture shows left lateral movement all
along its leugth, and the trace it makes on the hill sides suggests
that it dips to the southeast under Bear Mountain at angles vary-
ing from vertical to 4!")''. Where the trace of this rupture makes
a sharp turn as it does at 39, and the tendency would be for a left
lateral fault to pull apart, it developed grabens and extensional
fracturing on an impressive scale, some of the cracks being wide
enough to admit a man and up to 10 feet deep. West of 30 the
trace runs up and down over ridges and was at times hard to fol-
low, but ea.st of 30, the left lateral offset attains several feet and
was clearly indicated by offset foot paths, wheel ruts, fences and
the sides of a turkey pen. The vertical offset was variable, the
south side being uplifted in some cases, the north in others. Where
the trace passes near the easternmost house of the Rogers Ranch,
a frame l)uilding occupied by Mr. and Mrs. C. V. Thompson, it
transects a fence, the posts of which were oft'set in an interesting
manner. The post directly above, and seated in the moletrack was
vertical, the posts on either side sloped away from the moletrack.
The trace of the fence was also surprising in that it simply bulged
or bowed a foot downhill in plan over a distance of (50 feet, to
the northeast, along the trace of the moletrack. There was no
apparent right or left lateral offset on this fence. The wires were
very tight and some were broken. There is no simple way to
explain this odd distortion, as the road is offset just northeast of
here and there is abundant evidence of a left lateral offset to the
southwest of this point. Oldham described a similar feature which

Part Tl

Geology

he cmMi'iI till' Itordwar fiactiirp on piiKe 14!) I't si'ij. of liis ropurt (in
tlic 1S!I7 Indimi carllKiiuikt'.

Tli<> ti-aoi' folks at 10, within 100 feet of this point and oni'
liiaiu'h, the liealville fault, departs from the Keneral zone of
fracturinc and strikes N. 10° E. as does tlie cross fault in the
rejjioii of the White Wolf Kanch. The other liraneli roughly ]iaral-
lels the fence just deserilied, and eventually readies the railway
tunnels which were damaged hy the earthipiake. The fact that this
trace runs parallel to the fence for a distance may have altered the
offset of the fence s(unewhat and perhaps is responsilile for the
peculiar fence displacement.

Ilcdlvilli' Fault. After branching off at point -10, the Bealville
fault passes through a tield and crosses the road at the intersection
of the Itealville road and highway 460; here it offset the fence in a
left lateral ninnner and proceeds as a moletrack around the north-
west side of the hill, locally known as "Shaking Mountain." It
then crosses the Bealville road, displacing fences on both sides of
it by about 1 foot and snapping fence wires. Where the trace
crossed the railway tracks 800 feet west of Bealville 8 inches of
rail were removed by workers to correct the shortening. The trace
crosses a field and ends at 41. a quarter of a mile north of the
railway tracks. In places the trace is marked by open cracks with
6 inch gaps ; in others the ground surface was humiied up. The
vertical movement varied, being alternately down on one side and
then on the other. Left lateral en echelon cracks mark the course
of this fracture.

South of Snirmill. Returning for a moment to the region south
of the sawmill, there are several rupture traces parallel to the
main moletrack. Near 4.3 is a spring, serving the sawmill, oiiened
by a tunnel driven into the mountain. The r<iof of the tunnel had
caved in about 20 feet from the entrance. A typical compressional
moletrack passes parallel to the front of the hill just lielow the
spring ; to the west it bifurcates, one branch going up the hill, the
other following the base. The upper trace sotui ends, but the lower
goes around the hill and up into a canyon where the upper trace
reappears and joins the lower trace. It then cro.sses the canyon
and enters it again as the canyon swings right. Beyond this point
the trace fades out but it reappears in a short distance and con-
tinues up the hill parallel to the longer track above it. The indica-
tion of movement on this feature is predominantly left lateral with
alternating compression and extension along it, but in general there
was more extension ; it, however, did not amount to more than a
few inches.

To the east, the trace has developed characteristics of a landslide
or slump feature and continues with interruptions until a trace
with a N. 2r>° E. trend was found at 43 leading eventually to the
Bealville fault which it joins. The predominant movement on both
was left lateral.

Railwny Tunnel Faults. Returning to point 40, the southerly
branch of the fault may be traced easterly. This branch swings off
within 100 feet of the bulged fence near the easternmost Rogers
Ranch building, runs subparallel to it for a matter of a few hun-
dred feet, and then strikes across a field displaying a fine set of
left lateral en echelon fractures. Where it crossed fences, posts
were tilted and wires broken. It then goes over a low ridge just
west of highway 466 without changing trend and crosses the high-

♦

i^/vfe - „,.

Figure 6. Normal faulting at point 44. This trace, or branch of
it, passed through tunnel 4. T. R. Fahy photo.

way un<ler a marker post designated KER.^BE. The pavement was
cracked and broken and the moletrack emerges on the northeast side
<if the highway as a fissure open 6 inches to a foot and displaying
signs of small left lateral movement. The northwest side of this
fissure went up. This means that the upthrow was, locally at lea.st,
on the uphill side and on the block north of the fault. These traces
on "Shaking Mountain" at 44 were visited by many people in the
weeks following the earthquake. The cracks continued to open after
the earthquake. They were first observed the morning following the
earth(|uake and then again 2 days later; in the intervening time
they had opened an additional ;{ or 4 inches.

As this trace continues eastwar<l it frays into several branches
which die out and into two important branches at 4.'>, which pass
over the brink of the hill above tunnels ,3 and 4 of the Southern
Pacific and Santa Fe railroads. One branch passes into a galley
just south of tunnel ',i and the other through tunnel 4.

Figure 7.

Detail of normal faulting near point 44.
T. R. Fahy photo.

Tunnel Area, The fault zone crossed the Southern Pacific
Railroad tracks like the bar in a dollar sign and the three tunnels
at two of these crossings suffered severely. Huge excavations and
fills were made immediately to reopen the railroad. The tunnel
offsets and the high and costly cut faces afforded the best informa-
tion to be found anywhere along the fault zone bearing on the
nature of the fault movement. Only here were cross sections of
the ruptures brought to view. These exposures shed considerable
light on the true nature of the ruptures or moletracks followed
on the surface for miles to the southwest and northeast. Much
more detailed information was gathered in the field than can be
set forth here.

The southern part of the 700-foot north-south Tunnel 3 was
so badly damaged that the southern 206 feet of it was converted
to open cut. At the south end the arch or upper part of the
tunnel moved relatively 10 inches .south with reference to the
lower part along a nearly horizontal fracture at the spring line.
This was shown by the offset of the portal face and by the bent
reinforcing steel. The lower part of the tunnel walls or lining
was shoved inward toward the center about 3 feet. The steel
rails of the single track were thrown into letter-S figures both
inside the south end and south of the tunnel and pushed sidewise
through the concrete lining to the rock walls. The deformation
of the rails has been described in another section of this bulletin
( Kupfer, Muessig, Smith, and White). The movement of the
crown portion of the tunnel with reference to the lower part and
the kinking of the steel rails indicate or strongly suggest hori-
zontal shortening in a north-south direction such as might result
from reverse movement on a southeast-dipping fault.

The clean-wared open cut made south of the south portal at 47
displays beautifully a reverse or thrust fault dipping 20-30 de-
grees southeast and striking about N. 45° E. It rises northward
from below track level on both the east and west sides of the
cut from a point about 100 feet south of the new portal to an
elevation slightly above the cap of the concrete portal face and
hence was exposed on three sides of the cut. On the east side
of the cut the rock above the thrust surface is shattered diorite ;
below it are beds of somewhat compacted sand and boulders

Earthquakes in Keen County, 1952

[Bull. 171

FliiURE 8. View west toward fault trace just east ot south portal
of tunnel 4. T. K. Fahy photo.

dipping about 3° northerly. They are probal)ly old Quaternary
sediments and not less than 3(1 feet thick. These brown and yellow
beds enclose the south end of the present concrete tunnel barrel.
Ten feet west of the portal face, before being covered by concrete,
they could be seen terniinatiug and abutting against a diorite
surface sloping 40' E. ; this was quite certainly not a fault but a
depositional contact. The trace of the thrust fault above tlie sedi-
ments can be followed just over the top of the tunuel; where the
sediments end west of the portal it enters diorite and descends
southward to track level on the west face of the cut. The diorite
above the thrust is badly shattered and is cut by numerous
northwest-trending steep minor faults, marked by gouge layers
J-i inch in width. The trace of the thrust dips about 30° at track
level but is convex upward and practically horizontal above the
tunnel portal. It is clearly an old fault, for there is commonly
half an inch of gouge along it, and in some places as much as
10 inches in pocket-like accumulations. In a cut 50 feet east of
and at the same elevation as the portal cap, the striations on the
fault surface strike about N. 35° E., suggesting mainly left
lateral displacement.

One of the interesting features of this thrust fault is that
while movement presumably occurred on it during the main or
July 21, 1952, shock which severely damaged the tunnel, which
it cut, displacement continued on it after the earthquake. The
cut was made about August 1 and a photograph of the west face
of the cut made on September 1, 1952. The hanging wall had
moved eastward over 2^ inches during August. On an unascer-
tained date, nails had been driven into the gouge above and below
the slip surface, their heads originally in contact. Their .separation
indicated that the direction of movement of the upper block was
about N. 45° E., or almost entirely strike slip with left lateral
displacement. From dated pencil marks on the underside of the
hanging wall it is clear that the movement did not occur at any
one time but was distriliuted. irregularly or regularly, between
August 1 and September 15. Oddly, the trace of the thrust fault
on the clean east face of the cut showed no offset what.soever.
This raises the question whether the post-earthquake offset cui
the west face of the cut was the result of aftershocks or fault
creep on the one hand or of merely settlement and plastic spread-
ing of the shattered rock in the hill mass above the fault, on the
other.

Tunnel 4, now abandoned, was a few hundred feet south of
Tunnel 3, trended northwest, and was 334 feet in length. The
tunnel wa.s so badly danuiged by fault offset, collapse of roof at
several places, and shattering of lining that the Southern Pacific
Company, instead of repairing it, cut a shelf at tunnel floor level
across the hill spur through which the tunnel passed, immedi-
ately east of the tunnel, and re-located its track on it. The
uncovered barrel of the abandoned tunnel remains. Rising above
it lo the southwest is a huge cut face, some 400 feet long, roughly
200 feet high, with perhaps 1:1 slope and several berms ; an

unfortunate necessity, it is a magnificient geological exposure in
the fault zone.

The most severe damage in the tunnel was about 80 feet from
the south portal, where a fault cro.s.'^ed it and caused uplift of
about 3 feet, and a shift of about 2 feet eastward, of the block
north of it. This fracture continues downhill to Clear Creek and
is probably the same break that is .so well exposed at 48, at the
south portal of Tunnel 5. Westward from the tunnel it rises
obliquely up the 200 foot face of the cut with a dip of about 30°
southward, and a strike roughly east. The gouge along it is as
much as .3 inches wide. The diorite above the fault is gray in
color, less shattered and weathered than the diorite below it,
which is brown and badly broken and decomposed. The crushed
zone along the fault is about 3 feet wide and contains good
spherical fault-rolled pebbles, "rollers," 1 inch-3 inches in diameter.
Striations on the footwall on the berm at the top of the tunnel
indicate dip slip movement ; on the next higher berm, 50 feet
above the tunnel, the striations .slope 30° eastward on the footwall,
suggesting mainly right lateral movement. Below the fault there
is a brown weathered zone some 5t) feet wide, parallel to it. and
northwest of it is more gray, less shattered and weathered diorite.
Viewed from^ a distance the brown weathere<l zone seems to be
steeper than the fault.

Tains /

trvei of roadbed

Figure 9. Thrust fault that cut tunnel 3 ; view east.
Sketched from photo.

The rupture or zone of ruptures which cros.ses Highway 466
at 44 forks on top of the hill at 45, above and west of the tunnels
and one branch goes down a gulch to the south end of the original
Tunnel 3 at 47, where so much damage was done. The more
southerly branch goes to the top of the 200-foot high face above
Tunnel 4 at 46, and after itself forking, connects with the fault
above described which goes through the south end of Tunnel 4.
We have the dilemma that the faults indicated at the tunnels
show displacements of at least several feet while the moletracks
which are presumalily their continnaticui on the hill above show
relatively small offsets both horizontally and vertically.

A slice of bedrock remaining along the northeast side of the
tunnel barrel consists of brown somewhat weathered iliorite cut
by a number of roughly east-west faults dipping southward
45-75°. One minor fault dips about 45° X. The main fault which
offset the tunnel is steep in this cross section, still strikes
approximately east, and shows al)out 3 feet of breccia and several
inches of brown gouge.

Below the track and roughly opposite the northwest end of
Tunnel 4 conspicuous rock outcrops existed, now largely buried
by fill. A striking fault cut this outcrop ; it dipped about 60° S.,
with strike of about N. 60° E. It has somewhat the same trend
as the main fault which cut the south end of Tunnel 4 but must
have been some 300-500 feet north of it.

Another fault cuts the upper part of the south end of the
200-foot cut face above Tunnel 4 ; it dips roughly 30° S. and
strikes approximately east-west.

Part I]

Geologt

FiGi'RE 10. Fault and gouge in west side of railroad cut. Hang-
ing-wall block (upper) probably moved upward during earthquake,
but displaced nail heads show downward landslide-type movement
following earthquake.

Longest of the tunnels (1169.6 feet), under the most cover
(over 200 feet), and through badly shattered rock. Tunnel 5 was
very severely damaged and required months for repair. It is on
the east or opposite side of Clear Creek from Tunnels 3 and 4,
is northeast of them, and hence also in the fault zone. Collapse
of the tunnel roof and failure of the disintegrated dioritic rock
resulted in three or four glory holes on the hill surface above
the tunnel. Long sections of the bore were filled with material
which flowed in from the roof. Mr. Jlehrwein reported that in one
section of ItM) feet in the tunnel the track was shortened 2.33 feet.

For train operation a shoofly was built around the end of the
spur pierced by the tunnel and along it a face over 1.0(K> feet
long was cut nearly normal to the fault zone, all in shattered
diorite. Only one fault was found cutting this face ; it is near
its south end. It dips about 4.5° S., and strikes about X. 75° E.
It is accompanied by a crushed zone about 1 foot wide, with
breccia and gray gouge. Traced eastward up the crest of the
spur this fracture probably connected with the southern of four
long cracks above the tunnel.

The cracks above the tunnel form a zone some hundreds of feet
wide trending X. 65° E. and therefore roughly at right angles to
the tunnel. They extend from somewhat west of the tunnel line for
many hundreds of feet northeastward across the spurs e.xtending
southward from the crest of the ridge pierced by the tunnel. The
cracks are most conspicuous in a large landslide basin just south
of the crest and east of the tunnel line. They are up to 12 inches
wide and roughly vertical, widest in thick soil, narrowest in thin
soil. They are crooked and showed very little horizontal displace-
ment. The total widening across the cracks must have been over
5 feet. The most northwesterly crack of the four showed left
lateral displacement in its southwestern portion, right lateral in
its northeastern part ; the southeast side was down about 12 inches.
The other three cracks showed downthrow of 6-10 inches on the
northwest side. There are numerous shorter cracks south of this
more conspicuous zone, both east and west of the tunnel ; one of
these, in rising up the ridges east of the landslide basin, showed
strong left lateral displacement.

A rather conspicuous fault crosses the south portal cut ; it dips
about 80° N. and strikes roughly east. Its trend is quite irregular.
It may well he the fault which crosses the south ixirtion of Tunnel
4. At Tunnel 5 it connects with a moletrack on the natural land
surface both east and west of the cut. Eastward the moletrack
goes half a mile to a saddle in the crest of the ridge and ends
near a striking old landslide basin. This is clearly an old fault ;
the gouge zone along it is quite wide. It continued to creep on both
sides of the portal cut after the earthquake.

Ground ruptures are virtually absent north of Tunnel ;"). Tun-
nel 0 was not badly damaged but the railroad company deemed it
advisable to convert it to open cut.

Around the curve to the east of Tunnel 6, on the slope into
Tehachapi Creek, at Cliff Siding, the track was on a fill resting on
the hillside. It was lengthened by a gap of about IS inches, with
shearing of the rail bolts. This was not due to faulting, however,

^^- %t^

FlGlRE 11. View west toward remains of tunnel 4, cro.ssed by
the White Wolf fault zone, and destroyed by the earthquake. The
right portion of the tunnel was elevated about 3 feet and moved
eastward approximately 1 foot with resi)ect to the left end. The
shear zone is of darker color and is marked by gouge and rounded
rock fragments.

Figure 12. Sketch from photo, figure 11.

but to shaking ; the fill convex to the northeast, slid northeastward
and downhill, several feet in places. There were no ground ruptures
hereabouts and Cliff was presumably somewhat northwest of the
zone of most acute deformation.

Tunnels 7 and S are a mile to a mile and a half to the southeast
and were only slightly damaged. They are apparently southeast of
the fault zone. Along the railroad between Cliff and these tunnels
there was virtually no ground rupturing and no evidence of sharp
deformation.

Resuming the discussion of the ground cracks and traces, the
hill lying between Clear Creek and Tehachapi Creek was fractured
in a complex and bewildering way. The cracks ha\'e the generalized
trend of the White Wolf fault zone, but are irregular in plan, both
on small and large scale. The predominant movement on these
features was that of northwest-southeast extension. When lateral
movement could be discerned it was usually left, but there are ex-
ceptions; there is occasional vertical displacement. The direction
of uplift is fairly consistent on any one crack but not necessarily
on two adjacent cracks. Frequently the downhill slide of a crack
will be uplifted from 0 inches to a foot. The cracks cut across fea-
tures of the topography without being influenced much by them,
.several cutting across guUeys and ridges without changing trend,
but at the s;ime time tending to favor small cols. This suggests
that they are old fractures. The cracks reveal that the underlying
rock is so badly crumbled that it resembles alluvium. The cracks
did not in general lie above the seriously damaged areas in Tun-
nel 5. There was S(mie offsetting of the tube in the tunnel, but
most of the damage resulted from collapse of the tunnel roof.

Earthquakes in Kern County, 1952

[Bull. 171

Hummnry oi data regarding ground rupture.

Trace or location

Trend

Dip

Length
(ft.)

Vertical

Extension or com-
pression

Lateral movement

Comments

(map numbers,
pl. 1)

Side
up

Amount

(ft.)

Sense

Amount
(ft.)

Sense

Amount

(ft.)

■ Evidence

San Emigdio R. —
Wheeler Ridge

N45 to 55° E

Steep

200
3.000±

1-4

E
E

0 to M

— -

Lurches, parallel to
contours.

Comanche Point

3 - . . .

N40E

N55-60E

N30W
N50E
N30E

SE30

500
300

200 ±
200
1,000

200-900

0-H
O-'i

o-y2

Small

o-'h

0-1

C
C

9
?

L
L

R
R

Small
Small
Small

En echelon
cracks

Crooked, 30' wide.

Alluvium.

Landslide crack?

7 -

Zone H mile wide.

Main JIolktrack, East of Arvin

N30E

N50±E

Low
Low
Low
Low

V
V

SE
SE

Small

3-4

2-3

Small

4'.'

7
9

c
c
c

c
c

Small

1-2

Feet

Several

6?
<1
3-5

R
R

•>

Small

Small
Small
Small

Small

En echelon
cracks

En echelon

cracks
En echelon

cracks

Fence
Roots

9. --

Zone up to 50 feet.

From here on to 14.

Overthrust zone.

12 ---

13 --

N45W
Curving

IS . . -

100' wide zone, soil

slip.
Soil slip accompanies

17 -.-

moletrack.

4-5' high ridges.

100' zone.

N45W±
N20E±

Fence pulled up

tight.
Fence pulled down.

Trace at Top White Wolf Grade

221-..

?
?

c
c

>t
>1

R
R

>1
>1

En echelon
cracks

Nothing between 19

and 22.
Nothing between 19

and 22

Cross Fault Near White Wolf Ranch House

24-25

NIOE

NIOE
NlOEi.

V?
?

26 ...

29 - - . .

1-lH

>l
<l
>l

En echelon

cracks; path

& fences.
Fence trace

splits.
Fences
Road, fences.
En echelon

cracks

Part I]

Trace or location

(niftp numbers,

pl. 1)

Geology

Sumwnry of data regarding ground rupture Continued.

Trend

Dip

Length
(ft.)

Vertical

Side
up

Amount
(ft.)

Extension or com-
pression

Sense

Amount
(ft.)

Lateral inoveinent

Sense

Amount
(ft.)

Evidence

Comment*

Hktwkkn White Wolf Ranch Axn C'i.kak Cukkk

al-
as.

34-35-
36

37-
38-

39-40-
40-41-
42-. _.

45-46..
48

Curved

N-S

N-S
N45E

N70W
N45-50E

NIOE
N4S-50E
N20E
N50E

E-W

SE?

SE
V

2,000

N
NW

H-1

H-1
0

E
E

1±

0-H

2-3

1±

Small?

^-1

SmaU

o-yi

I
L

Small
Small

Izfc

>1
>1
>1

l=fc

Small
Small
Small

En echelon
cracks

En echelon

cracks

En echelon

cracks
En echelon

cracks

Landslide?

Fractures in road,
Allen's.

Paths, fences, en
echelon cracks.

Fences, rails, en
echelon cracks.

Joins 40 to 41,
fences.

Between Tehachapi and Caliente Creeks

49 -

N40E
S80E
N80E
N80E
N15E

N30W
N40W
N10-20E

N70E
N30E

N40W

S75°

1,500

700

3,000

2,500

500
500 ±
3-4,000

1,000 ±
2,500 ±

500 ±
400 ±

E
S
S

N
S
E

E
E

3-4

H
H
l-lH

a-H

O-H

c
c
c

E
E

Small
foot±

1±
2 to 4

I
I

I
I
L

2-3

0
0

Fence

En echelon
cracks

50 ---

56 - -

57 --

59-60

over 200' zone.

Column
Dip

Extension or compression
Lateral moTement

LEGEND:

Symbol

Meaning

Vertical

E
C

E.itension

Compression

Indeterminate

L
R

Left lateral
Right lateral
Indeterminate

Earthquakes in Keen County, 1952

[Bull. 171

Tunnel 5 has been the scene of considerable trouble in the past.
The roof had collapsed before, following a derailment and fire in
the tunnel. The rock flowed down producing a glory hole high
on the hill. Four of these glory holes revealed a foot of brown soil
and several feet of badly weathered diorite, grading into fresher,
but crushed rock.

Between Tehachapi and Caliente Creeks. Some of the largest
cracks and fissures produced by the earthquake were found on the
ridge separating Tehachapi and Caliente creeks.

At 49 a trace with a trend N. 40° E. went up the stream bluff.
The cast side of this fracture was elevated 20 inches. The crack
crosses the stream.

A compressional moletrack at ."lO ran N. 80° E. for at least 700
feet ; the south side was elevated 2 feet. In crossing an east-west
fence, the moletrack slackened the wires about a foot, indicating a
combination right lateral and compressional movement.

Beginning at ,^1 and trending X. 80° E. is a strong compressional
crack. The south side was elevated one foot and the trace when
crossing a canyon indicates that the feature dips to the south
75°. There seemed to be little if any horizontal motion and the
feature cuts across spurs and swales independently of topography.

The rupture at 32 was compressional and trends N. 80° E.

Farther up Caliente Creek at 53 an exposed fault dips 80° SE.
strikes N. 40° E. It appears from the exposure on the stream bluff
that the igneous bedrock crops out on the upstream side and the
sedimentary material on the lower side. The rock is so badly
macerated, however, that a decision was not possible.

There is a gigantic crack at 54 trending N. 15° E., the east side
of which was uplifted 3 to 4 feet. The crack was extensional and
at places developed a graben 50 feet wide and 4 feet deep. This
crack itself seemed to show no horizontal displacement, but associ-
ated en echelon cracks indicated a left lateral habit.

Figure 13. View east toward grouu ! i i;
tare between Caliente and Tehachapi Creeks.
Surfaces are jagged, demonstrating lack of ap-
preciable strike-slip movement.

At 55 is a crack a few hundred feet long, downthrown 6 inches
on the south side and exhibiting no lateral movement. Parallel to
the crack at 55 is a similar one at 56. This is situated on the crest
of the ridge and like 55 passes through a col between two knobs.
Its trend is N. 40° W., the south side was elevated 4 inches and
there was no indication of lateral movement over the several
hundred feet of its length.

One of the largest displacements found during the field work was
the big crack at 57 which crosses the ridge obliquely and ends on
the north side. In its niidlength and near the top of the ridge,
it is 24 to 40 inches wide. 6 feet deep ; the east side was up-
lifted 12 to 18 inches and there was about 6 inches of left lateral
movement. Northward this feature runs parallel to the contours,
passing through a saddle at a point where the slope of ridge sud-

denly steepens. The material downhill from it appears to be
sedimentary, that uphill appears to be igneous, but it is difficult to
decide.

A moletrack at 58 was uplifted C to 10 inches on the uphill or
southeast side, showed some left hand en echelon cracking and
trends N. 70° E. for a considerable distance, about 100 feet east
of the ridge.

There are two parallel cracks at 59, each 200 feet long, con-
sisting of beautifully developed shear patterns showing left lateral
displacement. The en echelon fractures are 3 inches wide, 20 feet
long, and trend north. The cracks at 5!) proceed intermittently to
the region of 60 where there is a zone 200 feet wide consisting of
about 9 cracks, all of left lateral en echelon habit and each having
a horizontal displacement between 2 and 6 inches. The total dis-
placement was perhaps 2 to 3 feet. There are in this zone some
parallel cracks each open about 2 inches. Some show uplift of 1 to
4 inches on the east or downhill side.

At 61 is a 4-inch vertical crack trending N. 40° AV. with no
vertical or horizontal movement.

At 62 there is a group of cracks, all extensional, with a fraction
of a foot vertical displacement and with no strike slip movement.

Between 62 and 6.3 the ground is broken by innumerable small
cracks. Southward from 63 the cracks decrease rapidly and no
significant break was visible between 63 and 64.

The fissures on this ridge were primarily extensional in nature,
but showed compression as they approached stream bottoms ; they
are disposed to pass through cols or saddles and yet run along the
sides of ridges and across canyons disregarding topographic fea-
tures ; they are long and comparatively straight. The trends seemed
to fall into two groups, one about N. 4,%° E.. the other more nearly
north-south. Almost all the lateral movement was left.

No large or continuous fractures were found to the northeast of
Caliente Creek, but there were some smaller ones.

Centennial Ridge. North of Caliente Creek and east of Harper
Canyon, the northwestern part of this ridge lies approximately on
the northeastern projection of the White Wolf fault zone. It
trends more nearly east-west than the fault. Harper Canyon,
straight in plan but crooked in detail, has the trend of the fault
and may well be an expression of part of it. although it appears
to be .somewhat northwest of its projection. Centennial Ridge was
examined for about 2 miles from its northwest end. Huge land-
slides which occurred during the main earthquake and its after-
shocks produced great scars on its lower south side and its west
end.

There are numerous ground ru(>tures along the crest of Cen-
tennial Ridge from its northwest end to about the 3.0()0-fiiot
elevation, near 65, a distance of about a mile and a half. They
are steep cracks striking from N. .50-80° E., and always roughly
parallel to the crest of the ridge. They are rather straight, not en
echelon, and showed no lateral movement. They were usually open
i-li inches, sometimes 2-4 inches. Where the ridge is rather sharp-
crested the cracks are in one zone along the top ; where the crest
is nearly flat there are usually two sets of cracks, one set near
each rim of the flat area where it drops off to the steep flanks.
Virtually no ruptures cross the ridge obliquely with the strike of
the fault zone. This fact, the tensional nature of the cracks, and
their location along the .sharp crest or along the rims of the flat
upper .surface of the ridge, lead to the inference that they are
primarily due to movement of soil down the slope during the
shaking rather than to faulting. However, their abundance and
their limitation to the portion of the ridge lying approximately
across the projection of the fault zone, suggest strongly that the
fault passes beneath the northwest end of the ridge.

From Centennial Ridge one could see scattered landslide cracks
along the north side of Harper Canyon and on hill slojies along
Caliente Creek to the east.

Harper Peak. Ground ruptures on Harper Peak (elevation
5,700 feet), about 10 miles northeast of the railroad tunnels, are
of interest becau.se they are the most northeasterly cracks found
and they are roughly on the northeasterly projection of the fault
zone. Mr. Weatherwax. a Walker Basin rancher who discovered
them, kindly drove the authors to them by jeep. They are on the
east and south sides of the top of the peak. Although curved they
strike about N. 50° E. Of the several cracks the largest was 1
inch wide and the northwest side was raised 1 inch to 2 inches,
and it continued for several hundred feet. There was no en echelon
pattern, and no suggestion of horizontal offset. The nests in which
individual boulders lie hereabouts show no enlargement and it is
clear the shaking was much less severe than at the railroad tun-
nels and at White Wolf Ranch.

Part I]

Geology

]\'(iUir llasin. Because the BreckenridKe fault, named by
I>il)lilee, wliirli created the imposing scarp west of Walker Basin,
has sonielinies lieen thought to he the continuation of the White
Wolf fault, this res;ion was examined carefully for Kround rup-
tures. Practically no grouml dislnrhance was noted along the Oiler
Canyon road into Walker Basin. 'Phe ridge leading eastward from
the summit on this road toward Harper I'eak displayed no cracks.

In a liorrow pit at U7. on the west side of the highway §
mile south of the Rankin Uauch, which is at the south end of
Walker Basin, several ruptures were found. This is at the soulh
end of the Walker Basin scarp. One about 12.j feet long crossed
the east edge of the quarry ; the south part of it is on a rock-cut
surface and the north part on an east-sloping grassy hillside. It
is crooked and a quarter to half an inch wide. About 400 feet
ea.st of the highway there is a crack about 40 feet long on the
top of the ne.\t little north-south ridge east of the quarry. It is
about i inch wide and trends N. .">° E. There are a number of
other small north-south cracks hereabouts. .\11 seemed to be tension
cracks: there was no suggestion of vertical or horizontal offset.

Xo ground ruptures attributable to the recent earthquake or
scarplets in the alluvium produced by geologically late movements
on the Breckcnridge fault were found at the base of the Walker
Basin scarp.

At the .Toe Walker mine, on the northeast side of Walker Basin,
a long irregular crack in soft wet earth marks the end of what
appears to he a landslide mass. A spring near the crack was flow-
ing vigorously on August .SI, 19.">2, 6 weeks after the earthquake,
and we were informed by Mr. Cannon that at the time of the
1946 Walker I'ass earthquake the discharge of this spring was
roughly quadrupled.

Breckenridge Mountain. This 7,000-foot north-.south ridge lies
ahmg the west side of Walker Basin and is apparently a block
tilted toward the west along the Breckenridge fault. On its south-
west slope a crack was formed at the time of the earthquake,
about A mile long and striking N. 40-60° W. It passes through
a col at the 4,400-foot level about 1 mile southeast of the junction
of Central Fork of Cottonwood Creek and Wei.ss Canyon, at 3.")"
24' 30" north latitude and 118° 36' .30" west longitude, on the
divide between Cottonwood and Walker Basin creeks. The crack
is about 4 inches wide, with downthrow of 4-") inches on the .south-
west side. The manner in which it crosses ridges and valleys
suggests that it dips 60-70° to the southwest. It seems to have
been displaced in a right lateral manner in some places, left lateral
in others. There is a landslide basin downhill from the crack in
some places, but not in others. There was a spring near the south-
east end of the crack. The rupture was di.scovered during range
riding by Mr. Charlton, who kindly led the authors to it at the
request of Mr. Leroy Rankin of Rankin's Ranch in Walker Basin.
Mr. Charlton reported that he did not notice any other cracks on
the west slopes of Breckenridge Mountain. This long crack is of
interest because it is about 0 miles due north of the White Wolf
fault zone at the railroad tunnels ; it presumably cuts bedrock and
is not merely a soil phenomenon ; it has apjiroximately the same
trend and seems to lie on the .southeastwaril projection of a line of
scarps extending northwestward from Allen Ranch through Hoosier
Flat to Kern River with a strike of N. 4.")° W.

Garlock Fault
On the day following the earthquake the senior author,
through the courtesy of Mr. Hearst of "White Oak Lodge,
examined the Garlock fault for about 18 miles, from
Cottonwood Creek we.st of the Lodge to Cameron. Nu-
merous short lurch cracks were found crossing the road
at different places and with various trends; they were
attributed to shaking. At one locality the ground rup-
tures seemed to have more significance. The paved Oak
Creek Pass road to Tehachapi, 0.8 mile northwest of its
junction with the Oak Creek road, is crossed nearly at
right angles by a zone of cracks; it is 4 feet wide and
the roadbed was dropped 6 inches between the two
outside cracks, necessitating a detour, regrading, and
repaving before the road could be put into use again.
This is exactly where the Garlock fault crosses the road.
The cracks extend 100 feet west and 300 feet east of the
road. Neither side was appreciablj- uplifted with ref-

erence to the other, nor could any lateral displacement
be discerned. The trend of the zone of cracks is that of
the fault. While the low and damp meadow west of the
road approaches a sagpond in form the area east of the
road does not appear to be deeply alluviated, so the area
should not be particularly susceptible to lurching. It
may be that the ruptures merely resulted from shaking,
but their length, their position exactly on the fault (and
yet no other long cracks found anywhere else in that
territory), and the coincidence in strike of the zone
of cracks and the faidt, cannot but cause one to suspect
that some slight local movement or other change in the
Garlock fault, presumably triggered by the main Arvin-
Tehachapi earthquake of July 21, 1952, may have pro-
duced the cracks.

Landslides

Landslides, a common phenomenon on steep slopes in
all strong earthquakes, developed on a huge scale in the
Arvin-Tehachapi disturbance and its aftershocks. There
are two aspects of this subject. One comprises the slides
that occurred during this earthquake; the other relates
to downhill mass movements of earlier decades and cen-
turies along the White Wolf fault zone.

There were many hundreds of large and small slides
on the morning of the main earthquake. They were of
course most numerous near the causative fault but many
occurred 50-60 miles from it. The main Los Angeles-Saii
Francisco highway, the Ridge Route (U.S. Route 99),
was blocked at a number of places between Grapevine
and Castaic. Large quantities of rock came down onto
the Pasadena-Vincent highway over the San Gabriel
Mountains. The road along Caliente Creek between Har-
per Canyon and Loraine was closed by rock slides for
weeks, as was the road up the Kern River gorge east of
Bakersfield. In nearly all the deeper canyons on the
northwest face of Bear Mountain slides occurred. The
steep slopes around Sycamore Canyon, and even some of
the gentler areas high on the mountain around the head
of this deep cleft, suffered severely and spectacularly
from landsliding. Canyons were dammed with rock debris
and some small lakes were formed.

FiGUKE 14. Ground ruptures forming small graben in trace of
Garlock fault. View westward across Oak Creek Pass road.

Earthquakes ix Kern County, 1952

[Bull. 171

Dependent upon topography and rook type the slides
took quite diverse forms. Some were types of slides found
both under ordinary conditions and after earthquakes:
rock falls; avalanches or rock slides; long but narrow
shallow soil flows; and old deep and massive landslides
which resumed movement for a few feet, opening up eracks
at their heads and buckling the ground at their toes. A
type unique to strong shocks consists of the movement
of the soil as a sheet over the bedrock over quite a large
area, sometimes several acres, with roughly subparallel
ruptiires distributed over the entire area. This was well
developed around Sycamore Canyon. In other cases the
soil sheet slid down one or both steep sides of a ridge
with tension cracks along the crest, or along the two
edges or rims of the crest where rounded or nearly flat.

Landsliding on the northwest face of Bear Mountain
continued for at least two months after the main shock,
probably mainly under the stimulus of aftershocks.
Whenever one of the numerous aftershocks was felt,
clouds of dust from landslides would be seen rising out
of the canyons shortly afterward.

The second aspect of the landsliding related to the
White Wolf fault is that along the whole lower north-
west face of Bear Jlountain and in the flat upland valley
lying northwest of and parallel to it landsliding on an
enormous scale has apparently been going on for cen-
turies in the past. A large part of a strip from half a
mile to a mile wide from Little Sycamore Canyon to the
railroad tunnels presents striking landslide topography. It
is quite certain that many or most of the small hills in
this zone, probably many of the large ones, are the tops
of landslide masses. A considerable area northwest of
the White Wolf grade, 5 to 9 miles east of Arvin, and
a much larger tract south of it, reaching up on the
mountain slopes and extending eastward to White Wolf
Ranch and beyond, shows convincing landslide topog-
raphy and macerated rock material. Equally striking
subsidence topography lies between the railroad tunnels
and the Tehachapi-Bakersfield highway. These landslide
masses are mostly large ones, up to hundreds of feet
long and wide. Many of them showed little or no effects
of movement during this earthquake period. Their unique
and characteristic features are that they form ridges or
long rounded hills that parallel the mountain front in-
stead of running down the slope as normal ridges between
canyons ; they often have steep faces toward the moun-
tain front as well as away from it; they often have
abnormallv flat depressions behind them on the side
toward the mountain, some depressions resemble or actu-
ally are closed basins; the ridges sometimes divert drain-
age so that it runs nearly parallel to the mountain face for
hundreds of feet; the topography as a whole is the
hummocky type so typical of landslide areas; and the
material of which the ridges and hummocks are made is
comiiletely shattered and much of it is a jumbled mass of
rock fragments and fine material.

Landslide topography is so widespread and so marked
along the northwest base of Bear Mountain that the
authors were very dubious during much of the field
investigation whether all of the ground ruptures traced
and mapped were not merely landslide features. Un-
questionably a large fraction of the total number are of
that origin", especially the curved and short ones and

the ruptures that trend in directions quite different from
the strike of the fault zone. But the long straight ones
trending northeast are in all probability the surface
expressions of branches of the WTiite Wolf fault that
experienced displacement at the time of the earthquake.
It would appear from the authors' observations that an
active reverse fault with numerous branches, creating a
high scarp, is a very favorable zone for landsliding on
a large scale. It creates a wide zone of crushed, pulver-
ized and jumbled rock readily amenable to weathering
and open to surface waters; the block above the fault is
shattered and weakened ; the fault movements produce
over-steepened slopes and a tendency to overhang by
repeated uplift of the scarp side of the fault; and vio-
lent shaking from time to time resulting from the move-
ments aids the constant downward pull of gravity. This
seems to the authors to be the explanation of the extreme
amount of landslide activity that has occurred along the
northwest lower portion of Bear Mountain.

Dislodged Boulders

At many places within a few miles from the fault
large boulders resting on hillsides were dislodged by the
earthquake and rolled down hill varying distances. In
one of the canyons on the face of Bear Mountain south
of White Wolf Ranch a sub-spherical boulder about 10
feet in diameter rolled down a long steep hillside, bound-
ing 200-;5()0 feet at a time and cutting trenches 2-3 feet
deep at each contact ; it finally stopped after mowing
down some quite large trees. On the north slope south
of the sharp switchback curve half a mile west of the
junction of the east-west Caliente road with the main
highway, at 71, several large boulders rolled down the
hill and one of them jumped the highway. All left spec-
tacular curved dribble paths. At an elevation of about
1400 feet on the White Wolf grade, at 72, a rock about
the size of an automobile rolled down against a highway
fill near a culvert. About 10 miles from the fault, on the
southeastern extremity of Bear Mountain, on the north-
east sides of Ciimmiiigs and Brites valleys and about a
mile and a half northeast of the former California Insti-
tution for Women, many rocks rolled down the hillside
into the canyons and left interesting dribble trails. This
is the greatest distance from the fault that extensive
rolling of boulders was noted. Landsliding seems to have
occurred at much greater distances from the fault than
the rolling of boulders.

It is interesting that, as might be expected, it was the
large boulders that rolled down the hillsides; the small
ones either were not dislodged or were soon trapped. The
smaller ones apparently could roll down only the steep-
est slopes.

Some large rock masses in outcrops or still resting in
their nests seemed to have been elevated a fraction of an
inch by the shaking and presumed rocking, which per-
mitted smaller rock fragments to roll or slide under
them.

In places witliin the White Wolf fault zone the shak-
ing apparently actually jostled some of the larger boul-
ders at least partly out of their nests, so that they were
i-otated a bit when they came to rest.

Within one or two hundred feet, and only at that
short distance, from any one of the long straight rup-
tures considered to be actual fault traces, boulders rest-

Part I]

Geology

ing in soil often enlarged their nests in horizontal diam-
eter by 5-10 percent. This appeared to be an inertia
effect rather than due to rocking, but it could be both.

INTERESTING OR UNIQUE FEATURES OF THE
FAULT AND EARTHQUAKE

While no two strong earthquakes are alike with refer-
ence to the nature of the shocks and the character of
the faulting which causes them, tlie Arvin-Tehachapi
earthquake and the "White Wolf fault presented some in-
teresting and unusual features when compared with
other California earthquakes and earthquake-producing
faults.

1. The shock was the strongest in southern California
in nearly a century — since the 1857 Fort Tejon earth-
quake, which occurred on the nearest portion of the San
Andreas fault.

2. The White Wolf fault on wliich the Arvin-Te-
hachapi earthquake occurred is surprisingly short for
a shock of this magnitude ; its known length is only
about 32 miles. However, the area of the fault surface is
in all probability large enough to make up for the short-
ness.

3. The majority of strong earthquakes which have oc-
curred west of the Sierra Nevada have originated in the
Coast Ranges west of the Great Valley, but this series
of shocks centered at the south end and along the east
side of the southern San Joaquin Valley.

4. The fault on which the main shock originated does
not trend west of north, like the San Andreas and the
other faults on which so many earlier strong earthquakes
have had their sources, but rather strikes at right angles
to the San Andreas — roughly northeast — and subpar-
allel to the Garlock fault.

5. The White Wolf fault is apparently not a typical
strike-slip fault like the vertical San Andreas fault or
the vertical Garlock fault, but is mainly a reverse fault,
or perliaps even a thrust fault, which has experienced a
very large vertical component of displacement in the
past.

6. While generally oblique-slip, the movement on the
fault in the Arvin-Tehachapi earthquake was apparently
more dip-slip than strike-slip ; it apparently differed
somewhat along the fault, and involved other complexi-
ties— all in contrast to the relatively simpler strike-slip
movement on the San Andreas fault during the 1906
San Francisco earthquake and the 1940 Imperial Valley
shock.

7. The maximum intensity of this earthquake, which
is related to the vigor of the shaking and therefore to
its destructiveness, seems to have been lower than usual
for a shock of this magnitude.

8. There is some reason to think that the intensity
was higher on the southeast side of the White Wolf fault
than on the northwest. To judge from the damage at
Tehachapi, the California Institution for Women, and
Monolith, the vigor of shaking was nearly or quite as
great in that territory as it was at Arvin on the oppo-
site, or northwestern, side of the fault but only about a
quarter the distance from it (4| miles). To be sure the
damage at Tehachapi was mainly to old buildings, but
those at the women's prison and at ]\Ionolith were not
weak structures ; also, Arvin is located on the deep allu-

vium of the San Joaquin Valley, in which the vigor of
shaking would expeetably be accentuated. A plausible
explanation for the unsymmetrieal distribution of in-
tensity on the two sides of the fault might be that it is
due to the fault's southeast dip, toward and under
the Tehachapi region. Perhaps the actual permanent dis-
placement of the initial fling enhanced the intensity on
the upper block.

9. The White Wolf fault, on which the main earth-
quake originated, was not regarded by geologists as one
of those more active faults of the state along which most
of our stronger shocks develop. It was recognized as a
young fault because of the age of the youngest (Upper
Tertiary) strata which it cuts and the high, bold, and
relatively little-dis.seeted scarp on the end of Bear Moun-
tain which it created ; but the fault lacked such evi-
dences of recent activity as fresh scarps in alluvium, old
moletracks, sagponds and fault trenches — so common
along the principal active Coast Ranges faults.

10. Though ea.st- and northeast-trending faults in
southern California have for a long time been recognized
as active, and though the Santa Barbara earthquake of
1925 presumably originated on an east-trending frac-
ture, most or nearly all of the historic strong shocks in
the western part of the state have come from the north-
west-trending faults ; these have strikingly restless fault
physiography, like the San Andreas. Geologists and seis-
mologists have come to expect that future strong shocks
will emanate from these long northwest faults. The
Arvin-Tehachapi earthquake should modify judgment
somewhat on this score. Apparently shocks must be ex-
pected in the future from faults not in the old orthodox
category. Strong shocks are likely to originate on rela-
tively short faults as well as on long ones; on east- and
northeast-trending fractures and perhaps still other
trends, as well as on the traditional northwest-southeast
Coast Ranges directions ; and on faults which do not ex-
hibit striking indications of recent activity and which
on the basis of other geologic considerations would not
be regarded as being active faults. Since moderately
strong earthquakes (Bakersfield August 22, 1952 and
others) have now occurred along the eastern margin of
the San Joaquin Valley and the floor of the valley is
known to be folded and faulted more or less like the
Coast Ranges to the west, it is clear that strong shocks
will not always be limited in future entirely to the Coast
Ranges. They maj' be expected from foci beneath the San
Joaquin Valley and probably from beneath the Sacra-
mento Valley. This probability takes on added impor-
tance because of the thickness of alluvium beneath these
valleys.

11. The earthquake did not develop a simple clean-cut
trace along the fault, like the strike-slip ruptures on the
San Andreas fault near San Francisco in 1906 and in
Imperial Valley in 1940, or like the dip-slip scarps on
the Sierra Xevada fault along the Alabama Hills in
Owens Valley in 1872 or the fault along the east side of
Pleasant Valley south of Winnemucca, Xevada, in 1915.
Instead, a complex pattern of ruptures in a zone along
the fault, a half mile or more in width, were formed.

12. The ruptures have quite different trends and the
displacements on them are in diverse directions. There
are quite long minor faults meeting and crossing the

Earthquakes ix Kerx County, 1952

[Bull. 171

zone of ruptures at angles approximating 45 degrees.
From these facts and from the results of U.S. Coast and
Survey re-triangulation and re-levelling, it appears that
the movement on the fault may have been quite com-
plicated.

13. While the zone of ruptures marking the north-
eastern "20-2-1 miles of the White Wolf fault ends south-
westward at the Tejon Hills and the alluvial surface of
the San Joaquin Valley between the Hills and Wheeler
Ridge, for some 12 miles, does not indicate the existence
or the location of the fault, it is a very interesting fact
that another series of ruptures appears on the higher
part of Wheeler Ridge. The epicenter of the main shock
has been located by Dr. Gutenljerg and Dr. Richter some-
what south of the highest part of the ridge. The east-
trending ridge is an anticlinal structure pushed north-
ward toward the San Joaquin Valley on a rather flat
south-dipping thrust fault whose trace would lie near
the north base of the ridge. Presumably the White Wolf
fault passes beneath the thrust. The ruptures in the
upper half of Wheeler Ridge occur where the White
Wolf fault would pass under it, and their trend is the
same as that of the White Wolf. Careful search and in-
quiry revealed no ruptures in any other parts of the
ridge. It would be an odd coincidence if the only rup-
tures on the ridge, occurring on the projection of the
White Wolf fault and parallel to it in trend, were not
related to it.

14. It is interesting and rather odd that, huge as the
vertical offsets have been on the White Wolf fault, it
does not appear to continue southwestward from Wheeler
Ridge, and the epicenter of the main shock, to the San
Andreas fault, a distance of some 15 miles, or less than
half of its own known length. It is also odd that the
surface geology does not more clearly indicate whether
the White Wolf fault turns gradually northward at its
northeast end near Caliente and merges into the south
end of the long Kern Canyon fault zone — into that en
echelon southern member of it mapped along the west
side of Walker Basin by Dibblee as the Breckenridge
fault — or whether the White Wolf continues northeast-

ward from Caliente and whether the Breckenridge does
not continue southward to intersect or join it. Possibly
the White Wolf connects with both the San Andreas and
the Breckenridge faults at depth.

15. The epicenter of the main shock is even southwest
of the known southwest end of the White Wolf fault and
from the fact that all the aftershocks occurred northeast
of the epicenter it is believed that the slip on the fault
that caused the earthquake progressed in only one direc-
tion from the point of initial rupture. Only the single
foreshock, which occurred about 2 hours before the main
shock, originated southwest of the main shock epicenter.

16. The many aftershocks, a number of them actually
moderateh' strong earthquakes, did not all originate in
the fault which caused the main shock; a large fraction
of them apparently had their source in the block above
and the block below the sloping fault surface and at a
distance of some miles from it.

17. From the initial shock at 4 :52 on the morning of
July 21 until the afternoon of July 22, the aftershocks
all originated in the block above and southeast of the
fault; later the aftershocks occurred in both blocks.

18. The sequence of events in connection with many
strong earthquakes has been thought to be the occurrence
of the main shock, preceded by some or no foreshocks,
and followed by a long train of aftershocks in or close
to the fault surface, decreasing in general both in fre-
quency and in magnitude during the ensuing months or
a few years. In the case of the Arvin-Tehachapi earth-
quake, in addition to the long train of aftershocks, a
series of quite independent earthquakes developed in the
months following the main shock, some of them presum-
ably on faults roughly at right angles to the White Wolf
and with epicenters up to 20 miles distant from it. Each
of these shocks had its own train of aftershocks. Some of
the shocks were moderately strong earthquakes which
did damage at nearby points, as for instance the one
which struck Bakerstield August 22, 1952, a month after
the Arvin-Tehachapi earthquake, and caused major dam-
age at Bakersfield.

6. GROUND FRACTURE PATTERNS IN THE SOUTHERN SAN JOAQUIN VALLEY
RESULTING FROM THE ARVIN-TEHACHAPI EARTHQUAKE

By Archer H. Warne •

Abstract. A series of fnint surface lines oliserved on nerial
photos in the vicinity of Biikersfiel<l have lont; been thought by the
writer to be a clue to the existence of a system of closely spaced
lateral faults traveisins the area in a northwesterly Oirection. The
remarkable broad parallelism of these surface features, together
with their highly interrupted aspect and their frequent looped
shapes, has led to much siieculation regarding their significance and
manner of origin. Although no positive relationship can be estab-
lished, it has long been assumed that the steep horizontally slicken-
sided fractures cored in widely scattered deep wells in this area,
were of the same trend and system as the surface lines.

The initial earthquake of July 21, 19.")2, produced in the Bakers-
field-Arvin area a nuiDber of surface features which compare so
favorably with the older series of lines that there can be little <loubt
regarding their identical manner of origin. It is believed that during
most of the Cenozoic there has been a recurrence of slight shifts
on an ancient system of basement faults, with individual adjust-
ments reflected at the ground surface as oriented shallow sloughs
and lateral offsets.

Introduction. As a result of tlie initial earthquake
shock on July 21, 1952 numerous small areas in the cot-
ton fields in the southern end of the San Joaquin Valley
were intensely fractured. A study of observed effects in
the portion of this area lying between Arvin and Bakers-
tield forms the basis for the first half of this report.
Attention was directed to these ground disturbances by
the appearance of cracks and slumps crossing roads and
highways in several dozen localities, often rendering
them impassable. Other effects related to ground-crack-
ing included failure of levees and reservoir embank-
ments, offsetting and breaking of concrete standpipes
and buried irrigation pipe, and dislocation of concrete
foundations of houses and other structures. Not only was
buried pipe widely damaged, but the ground in many
fields was so badly fractured that water, when finally
obtained, could not be prevented from entirely disap-
pearing into the cracks crossing the cultivated rows.

Although the areas where earthcjuake fracturing was
observed are somewhat irregularly scattered through the
Bakersfield-Arvin area, they show a tendency to fall into
several belts trending in a northwesterly direction. De-
spite the large number of varied and often strong after-
shocks occurring over a period of many days following
July 21st, the fracturing in the flat area (fig. 12) is
believed to have been entirely the result of the initial
and strongest earthquake.

Jones Ranch Area. A farm house located about 5
miles southwest of Arvin unluckily happened to lie
within one of the fractured areas. Although the wooden
structure of the house remained standing, the foundation
was broken and offset in a number of places, and the
concrete walls of the small square basement were pushed

• Geologist, Richfield Oil Corporation. Manuscript submitted Febru-
ary. 1953.
Thanks and appreciation are extended for the generous assistance
of many friends and associates, including particularly the fol-
lowing : to Mason Hill, who gave critical suggestions ; to Rollin
Eekis, Irving Schwade, J. W. Mathews, and Marie Clark, who
made helpful suggestions \ to Elmer Marliave, who furnished all
air photos made since the earthquake, and to Clifton Johnson,
who also assisted in obtaining them ; to Edward M. Bien, who
loaned photographs ; to Ray Arnette and R. L. Bowman, who
assisted in obtaining additional photographs; to W'arren Stod-
dard and Gordon Dolton. who gave valuable cooperation in
mapping of ground fractures. Acknowledgment is tendered to
the Richfield Oil Corporation for permission to publish this
report.

to the shape of a parallelogram 10° (degrees) off square.
The ground cracks seemed to approach the house from
all directions, some showing a vertical offset of up to 1
foot, and open as much as 6 inches, while others ap-
peared merely as belts of cracks which opened only
slightly.

During an inspection tour of this area a set of these
fractures was found which lay in an open 80 acre field
immediatel.v to the southeast of the farm house just
described. The field contained no cotton crop and there-
fore offered an unobstructed view of the pattern as-
sumed bj- a group of fractures. The fractures were found
to be preserved clearly and in great variety of trend and
spacing. They were mapped July 27-30 by subdividing
the field into square plots of 100 feet on a side and
sketching in the individual cracks in each.

On the map (fig. 4) each individual fracture is shown,
the heavier lines representing those on which there has
been more than 6 inches of vertical offset. The soil in
this field was loose, silty, and sandy, but was sufficiently
coherent to retain a clear record of the intense ground
dislocation. The cracks in the predominating looser soil
areas were V-shaped in profile, indicating that a wedge
of loose material had disappeared into the space created
when the crack developed. Scarps, also generally beveled,
were occasionally sharp where irrigation had earlier
formed surface mud.

The zone of surface rupture exposed in this open field
continued into cotton fields for some distance to the
northwest and southeast of this locality. Their mapping
was prevented or made impractical by both lack of avail-
able time and by poor exposure due to the presence of
rows of cotton reaching 6 inches to 3 feet in height. The
presence of hardened surface mud, or even of a packed
dry dirt road cake seems to have had a marked effect
on both spacing and in some cases direction of the sur-
face fractures.

It was found that almost without exception each group
or belt of fractures had resulted in the formation of a
shallow depression, without any perceptible uplift of
its margins. The fracture patterns seem at first to consist
chiefly of hooked shapes having little or no consistent
direction. It may readily be seen that those groups not
actually constituting part of a hooked shape show a
predominant northwesterly trend.

Several surface profiles (fig. 6) were measured during
mapping, where maximum slumps of approximately a
foot and a half were found. These individual fractures
have a nearly vertical attitude and show no offsets or
offsets smaller than 1 foot. There is no noticeable
viplift around the fracture groups or in any part of the
area shown on this map (fig. 5). The fracture depres-
sions being distinctly limited, both in lateral as well as
longitudinal extent, it is not difficult to perceive that, if
left undisturbed, they would without a doubt eventually
become shallow sloughs.

The development of the thick swampy vegetation which
characterizes undrained sloughs woulcl inevitably create
a thick dark soil body which would stand out in contrast

(57 )

Earthquakes in Kern County, 1952

[Bull. 171

FiQUBE 1. Typical group of linear surface markings.

COALI NGA

\, Lost Hills

Terra Bella

^ Kern-

X -VILLE

■^

X :--^

Bakersfi eld

TehachapT

AFTER JENKI NS

FlQURE 2. Fault pattern in southern San Joaquin Valley area.

Fart n

Geology

^ COALI NGA

Terra Bella

^^^^ / Kern-

KERSFI E^D

TehachapI

Figure 3. Location of linear surface markings in relation to fault pattern in southern San Joaquin Valley area.

'W, X\ ri

^^t

v..

^^>-

Figure 4. Map showing linear surface markings in th« southern San Joaquin Valley.

Earthquakes in Kern County, 1952

fBiiU. ni

0 200FT.

Figure 5. Generalized outlines of groups of fractures.

APPROX. I^FT. MAXIMUM SLUMP

.,B'

^ 5FT.
=- 0

Figure 6. Cross sections A-A' and B-B' from figure 5.

Part I]

Geology

INITIAL EARTHQUAKE SLUMP

A^ / ^^

A >v \ ^-

SHALLOW SLOUGH 500± YRS

SOIL PROFILE AFTER LEVELING

Figure 7. Process of development of sloughs from fracture groups.

Figure 8. Jones Ranch group of fractures in relation to a square-mile grid.

Earthquakes in Kern County, 1952

[Bull. 171

CHEVALI
RD.

FiGUBE 9. Fracture zones crossing U. S. Highway 99 and Chevalier Road.

~,^-M 'III'

XUU\ ^OQ^ V/

C^Kr

Bakers- ^\

ECR^

Fl ELD

<i>

O r

0 SMI

1 I I I I I

1952 <;^^ \

FRACTURES \

'^V Arvin

# ::

WARNE

FiouBE 10. Relationship of 1952 fractures to ancient fracture pattern.

Part 11

Geology

Rio
Bravo

OIL <^^V^xx\S^

FIELD ^^ ^-^

^ Bakersfield

0 lOMI

WARNE

Figure 11. Hypothetical primary joint pattern relative to movements on Garlock and San Andreas faults.

BAKERSF I ELD

RIVER
Ea

A RV I N

R26E

■•v..

Figure 12. Belts of earthquake fracturing in the Bakersfield area.

Earthquakes in Kern County, 1952

[Bull. 171

Figure 13. Intense fracturing in iirit'iittd culton (itid.s.
Photo by E. M. Bien.

to adjoining soils upon removal of the slough or its site
by later grading and plowing.

U. S. Highway 99 and Chevalier Road. Slumps which
suddenly appeared across U. S. Highway 99, 11 miles
south of Bakersfield, at the time of the first shock inter-
rupted bus, truck, and tourist traffic. Two areas on that
highway, where the pavement dropped nearly a foot,
are only a part of a group of intense fracture zones. The
recent breaks lie within the areas bounded by solid
lines (fig. 9). The east wing of the school building in
section 18 was lowered nearly a foot by the slump out-
lined nearest to it.

The break in Chevalier Road showed a horizontal
offset of about 1 foot, combined with over 1 foot of
slump development. A most interesting and signifieant
fact is that this offset is exactly aligned with a much
earlier and perhaps ancient tree-filled slough extending
for half a mile to the southeast. Shown in dotted lines
(fig. 9), the former full extent of this .slough, as mapped
by the U. S. Geological Survey in the early 1930 's,
included a one-time distinct hook-shaped slough in sec-
tion 29. The remaining dotted outlines are in each ease
the position of former sloughs present at the time the
topography was mapped, but since then leveled out and
planted in cotton and alfalfa.

It is of interest to note that 7 months after their
formation several of the road slumps are still growing
deeper. This is true of those crossing U. S. Highway 99
at the locality just described, and also one on Shaffer
Road and one on McKittrick Road. This may be mainly
due to the weight of vehicles.

Ancient Linear Surface Markings. These observations
and events take on special significance when viewed in
the light of what has been learned in the past decade
from a study of ancient linear surface markings in the
Bakersfield area (fig. 10). The discovery on aerial photos
in the middle 1930 's of some rather indistinct dark lines
traversing plowed fields in the Bakersfield area led to
the conclusion that these were possible surface traces
of faults. The lines seemed to be rather broad for single
fault traces, and appeared much too interrupted along
their courses. But they could be seen to have a pre-

FlQURE 14. Fracturing alonf road marking .south border of
mapped field. I'hoto by E. M. Bien.

dominant northwesterly trend, and were found to lie
roughly parallel to one another.

The examination of several hundred aerial photos cov-
ering an area extending from Wasco and Porterville to
Elk Hills, San Emidio Ranch and Arvin revealed hun-
dreds of such markings. The lines were so faint that their
detection in most eases depended on careful study. When
the lines are all plotted on a regional luap the}' show
a broad parallelism, and it is easily seen that they must
belong to some sort of a system. The extreme faintness
of the lines, as well as their interrupted habit, has
probably been largely responsible for their having been
rarely recognized or for having been considered of no
significance. Generally there is little more clue to their
position than a line or band of contrasting soil colors
on newl}^ plowed farm land. The dots in the illustration
(fig. 1) represent in each case the darker side of a line
of color contrast.

Speculation as to what might have been the manner
of origin of these lines led to a consideration of old wagon
roads, ancient game or livestock trails, and even suc-
cessive old lake shore lines, but each failed to provide
an adequate explanation. The frequent appearance of
hooked or looped shapes onlj^ served to increase the
mystery of their origin.

FliiUUK 1.'

Hncklini; of road.s in Arviu area. Photo by
E. M. Bien.

Part I]

Geology

FIUUKE 10. Detail of intense fracturing along n.ail marking .smith
border of mapped fields. Photo by E. M. Bien.

It is noteworthy that these lines may have been recently
anticipated without their presence having: been known.
In the paper Strucfural Relation of Tehaciiapi Moun-
tains to the Sierr-a Nevada and the Coast Ranges pre-
sented by J. P. Biiwalda in 1946 at the Berkeley meet-
ingr of the Geological Society of America, a series of
northwest-trending: fractures in the Tehachapi Moun-
tains was described and their relation to the well-known
southern San Joaquin Valley structures suggested. In
this paper it was indicated that these Tehachapi frac-
tures are only a part of a great system of faults and
folds which extends beneath the San Joaquin Valley
floor across to the Coast Ranges beyond Coalinga.

The index map (fig. 3) serves as a key to the general
location of the system of markings found on aerial photos
and shows their relation to well-known outcropping
faults in the southern San Joaquin Valley area. The
surface markings fall approximately within a 15 by 35
mile rectangle enclosing the city of Bakersfield. It must
be emphasized that the lines on the index are a greatly
simplified version of the detailed map, and offer no
idea of the actual number and close spacing of the lines.

Relation of the MarJcings to Lateral Faulting. The
appearance of these lines on a map and their parallelism

FlGURF, 17. t'lievalier Road slump, which has a measured 1-foot
right lateral offset (note rim marks). Photo by E. it. Bien.

FiGt'RE 18. Ground fractures in cnttmi fields near .\rvin.

with regional fault trends as well as within their own
group immediately suggested lateral faulting. This fact
combined with the frequent appearance of lateral fault
evidence cored in wells led to a review of all the core
descriptions from wells located in the area of interest.
This survey revealed a large number and wide distri-
bution of cored steep fractures bearing horizontal slick-
ensides. Several incomplete and rather detailed subsur-
face electric log studies of oil-field-structure problem
areas have yielded strong evidence of a close-spaced lat-
eral fault system in older formations.*

It is also found that most evidence of lateral slippage
comes from cores of strata underlying the upper Mio-
cene, and that this evidence usually shows a downward
increase in abundance. This may represent the accumu-
lation of lateral shifting in older beds, to be expected if
adjustments have occurred throughout the Tertiary.

In an early attempt to prove that the surface mark-
ings were the result of lateral faulting, a trench in the
Famoso area was excavated 6 feet in depth and 30 feet
in length across one of the better defined surface lines.
The result of this effort was negative, and consequently
the significance of the surface markings remained ob-
scure until the 1952 earthquake. No proof turned up
during a ten-year interval that the surface lines and the
subsurface lateral slippage planes were of the same
origin and trend, or that they had anything to do with
one another. The map (fig. 10) of distribution of the
old surface markings shows the relationship of the lines
seen in aerial photos to the two described groups of
earthquake fractures, and the striking parallelism is at
once evident. The outlines shown here and the lines
within them are again extremely simplified and general-
ized.

The apparent absence of the surface markings in the
alluvial fan areas of the Kern River and the Caliente
and Poso Creeks led to extensive re-checking of photos,
but was finally accepted as a fact. It is readily seen on
the map that this absence is simply due to removal by
flood processes of features once present there. The great
number of ancient lines found in the areas protected
from effects of stream and river floods, when compared

• Unpublished report, Evidence oj Extensive Lateral Faulting in
the Bakersfield Area, Kern Co\inty, California, by A. H. Warne ;
60 pp., with critical summary by Mason L. Hill. 1945.

Earthquakes in Kern County, 1952

[Bull. 171

tiniitmiiiJI

Flc.iRE 19. Lurih cracks in irrigated field near Arvin.

with the relatively few potential features formed during
the major 1952 shock suggests that this is a ciunulative
record of many earthquakes like the recent one. The sig-
nificance of this gap in the surface features depends to a
large extent upon the time involved in the processes of
formation of the individual lines. It could be taken as an
indication that the shifting and fracturing forces have
been inactive for sufficient time to permit seasonal floods
to remove all traces resulting from the last period of ac-
tivity. This would suggest that perhaps the normal fre-
quency of recurrence of shocks causing surface fractur-
ing is actually once in several hundred years and that
we have only inhabited this area long enough to experi-
ence one. On tiie other hand, the extent to which the old
features resemble one another would lead us to believe
that they were all formed in a relatively short time, and
that after a quiet period, somewhat greater than the two
centuries of local recorded history, another cycle of fre-
quent adjustments has commenced. In either case we
may be fairly certain that these slump-forming adjust-
ments are not the first of their kind in this area, even
though no such events are known from the local recorded
history.

In considering the significance of the looped and
hooked shapes seen in fractiire patterns (figs. 1, 4, 5) it
is important that all the small component fractures, with
rare exceptions, show purely vertical ofi'setting or ten-
sional opening. Their appearance therefore suggests ini-
tial lurching and subsequent differential resettling to
form the commonly observed slumps. The confinement
of these slump features to belts having a definite trend
and the occasional appearance of strong lateral shifting

FlQliRE 20. Lurch oracl<s in irrigated field.

such as the Chevalier Road offset suggest that lateral
movements at great depth, and possibly considerable
age, have been transmitted as a variety of strains set up
in the shallow less consolidated strata. Activated by a
shock wave from a nearby earthquake, a number of
these strains would then convert to very local and varied
adjustments, some lateral, but each, in turn, functioning
as a small earthquake center, and producing its own
effects at the surface. The fact that the looped patterns
are independent and may even be seen to intersect in
places, indicates that they are the effect of energy de-
rived from various secondary sources lying in a belt
rather than from a single point or along a single fault
plane.

Much of the subsurface structure encountered in
studying the oil fields in the area lying to the east and
southeast of Bakersfield suggests, in addition to the
common normal faulting, a set of northwest-trending
steep faults having a varying degree of lateral move-
ment, and intersected by a similar but lesser northeast-
erly set. In the final illustration (fig. 11) a joint system
was drawn representing a hypothetical primary joint-
ing of the regional basement rocks as a result of the
overall initial strain set up by early movements on the
San Andreas and Garlock faults. Subsequent movements,
both horizontal and vertical, on this set of basement
joints could easily account for much of the structural
complexity observed, for example, in the Mountain
View and Edison oil fields, as well as in the Racetrack
Hill and Ant Hill oil fields. The influence of such a
fault system is suspected in a dozen other oil fields lying
in an area of wider radius.

7. ARVIN-TEHACHAPI EARTHQUAKE DAMAGE ALONG THE SOUTHERN PACIFIC
RAILROAD NEAR BEALVILLE, CALIFORNIA!

By Donald H. Kuifek,* Sieukkikij Muk.ssig,' Gkokge I. Smith.' and Georgb N. White*

ABSTRACT

The ArviiiTehachiipi earthquake occurred iu south-central Cali-
fornia on July 21. li).")^, on the White AVolf fault. Where the fault
zone crosses the Southern Pa<ilic Hailroad. four tunnels were
destroyed, rails were twisted and Inickled, and in one area about
10 feet of crnstal sliortenini; was measured. The tyiies of (lainage
associated with the earthi|uake, and their distribution relative to
each other, seem to have been caused by movement iu a reverse- or
thrust-fault zone that dips .south. The damage resulted from com-
pression and from subse<iuent relaxation along normal faults.

INTRODUCTION

Surface displ;K'einent durinji; the Arvin-Tehachapi
earthquake of July 21. 1952 took place principally aloug
the White Wolf fault, the trace of which lies along the
base of a pronounced escarpment that forms the north-
west slope of Bear ilountaiii. Kern County, California
(California Division of Mines, 1952; Benioft', et al.,
1952; Buwalda, 1952). Severe damage occurred wliere a
fault zone, presumably an extension of the White Wolf
fault, crosses the Southern Pacific Railroad tracks 1,500
feet southeast of Bealville Railroad Station. Before the
earthquake the tracks here made an S-shaped curve and
passed through four tuiniels; all three limbs of the curve
and three of the four tunnels were intersected by the
fault zone.

The authors visited the area on the day of the earth-
quake. On that day and the two days following, they ex-
amined the faidt trace at several points between Beal-
ville and the bend in the road east of Arvin.

As bulldozers were about to destroy much of the evi-
dence of earth movement along the railroad, the plane-
table map was made by the authors on July 22, 1952. On
August 14 and 15 Kupfer and Smith revisited the area
and examined the damage in tunnel 5, which had been
impassable during the previous visit. The purpose of
this report is to present the authors' observations — most
of which can not be made again — and their conclusions.

Ack)wwledgmenis. The authors wish to thank Mr.
D. J. Russell, president of the Southern Pacific Railroad
Co., for the courte.sy and cooperation that he and his
men extended during the investigation; Messrs. E. E.
Earl, G. F. .Mchrwein, D. P. Boykin, F. M. Misch, W.
Jaekle, W. E. Bussey, and S. T. Moore, all officials of the
company, were particularly helpful. The Kern County
Land Co. loaned the plane-table equipment.

Terminology. The tunnels on the Southern Pacific
Railroad are numbered consecutively from Bakersfield
to Mojave. In local railroad parlance, the directions on
the track are referred to as "west" toward Baker.sfield
and "east" toward IMojave, without regard to actual
compass direction. This termiiiology is used in this re-
port only when referring to location of tunnel portals.
Distances are given from the west portals of tunnels,
and "left" and "right" refe.' to the left and right side
of a train heading toward Mojave or Los Angeles. Figure
1 shows the general relations of the area att'ected ; figure

• Geologist, U. S. Geological Survey, Claremont, California.

t Publication authorized by the Director, U. S. Geological Survey.

2 represents in more detail two of the tunnels and the
part of the tracks that were disturbed; and the photo-
graph in figure 4 shows their relation to local topog-
rapliy.

GENERAL GEOLOGY

In the area of the railroad tracks and tunnels, the bed-
rock is predominantly granular intrusive rock, cut by
small pegmatite dikes. Xo microscopic examination of the
rock was made, but it seems to range in composition from
a quartz diorite to a gabbro. It has been altered and de-
composed to an undetermined depth, and bulldozers were
therefore able to make sloping cuts 100 feet or more into
the bedrock without the use of explosives.

During the reconstruction of tunnel 3, an exposure of
nonintrusive bedrock, which is now partly concealed by
the finished tunnel, was found. It is a wedge of arkosic
material that was_ exposed from the arch of the tunnel
to a point 50 feet along the left (east) embankment. Most
of this rock is unbedded, compact, and fine-grained, but
it contains a few rounded cobbles up to a foot in di-
ameter. The top of this mass is in horizontal contact with
intrusive rock; in some places the contact is indistinct,
in others it is sharply defined by a zone of gouge 1 inch
to 2 inches thick. The lower contact was not exposed. No
further information was collected about this rock and it
is now concealed. For this reason its structural relations
are doubtful.

DISPLACEMENT ALONG THE WHITE WOLF FAULT
DURING THE EARTHQUAKE

The general trend of the trace of the White Wolf
fault is N. 55° E., subparallel to the Garlock fault,
which lies 18 miles southeast. The trace of the recent
offset was generally represented by several minor but
conspicuous fractures that cut the surface in en echelon,
parallel, or braided patterns along a zone up to 1,000
feet in width. The relative displacement of the surface
along most of these fractures was less than a foot, but
locally was up to 4 feet. Fractures of the normal, re-
verse, thrust, and strike-slip types were observed.

Along the base of the Bear Mountain scarp due east
of Arvin the movements apparently took place on a
thrust fault ; the southeast block was thrust relatively
over the northwest block. Two tear faults were observed.

In the vicinity of U. S. Highway 466 and the railroad
the surface fractures appeared to be on normal faults
with downthrow on the southeast. The average displace-
ment was 6 to 8 inches and included a strike-slip com-
ponent.

DAMAGE ALONG THE SOUTHERN PACIFIC
RAILROAD TRACKS

The railroad tracks between Bakersfield and Mojave
were damaged by the earth(|uake iu many places. Boul-
ders and large rock masses slumped onto the tracks and
most of the larger fills settled slightly, so that much of
the track had to be cleared, leveled, and straightened.
In the S-curve area near Bealville, the track and tunnels

(67)

Earthquakes ix Kern County, 1952

[Bull. 171

Miles

Figure 1. Trace of White Wolf fault, partially from data from T. W. Diblilee Jr. (this bulletin). Location of Map of railroad route
east of Beahnlle (fig. 2) is shown by solid rectangle. Topography from U. S. Geological Survey C'aliente quadrangle. Contour interval 1000
feet. Datum mean sea level.

were SO sevei-ely damaged that all tratSe on the line was
suspended for 25 days.

The tunnels in this area are lined with steel-reinforced
concrete walls from 12 to 24 inches thick. Large slabs
of concrete that broke away from all other adjacent
support were held in place by the reinforcing steel for
several days, but continued aftershocks finally broke
them loose.

Tunnel 3. Tunnel 3, originally 700 feet long, actually
runs a little west of south from the "west" portal. It
was undamaged from this portal to a point 548 feet
south, where a displacement approximately at right
angles to the centerline of the tunnel fractured both
walls from floor to arch. When examined on the morning
of July 22, both concrete walls of the tunnel were dis-
placed about 2 feet horizontally, the south side to the
east. By the time the plane-table map was made that
afternoon, the left (east) wall had collapsed inward.
From this fracture to the "east" portal, a distance of
152 feet, the walls of the tunnel were broken and large
slabs of concrete were loosened but kept from falling by
the steel reinforcement. In the last 70 to 90 feet of the
tunnel, the concrete was thoroughly shattered ; the arch
was broken, and the tunnel was caved in.

In repairing this tunnel, 206 feet of the damaged end
was converted into an open cut, or "daylighted." After
daylighting, the west wall of the new cut was examined
and two zones of broken and crushed rock were observed,
one 605 feet and the other 620 feet from the west portal.

At a point 570 feet from the west portal (76 feet south
of the new east portal) a fracture was observed on
which movement had occurred after the daylighting,
and by August 15 this movement had amounted to about
an inch. On November 1, when Smith visited the area,
the displacement had increased to several inches. This
fracture strikes N. 15° E. and dips 35° SE. at waist
height but flattens out upward. At the top of the new
tunnel portal the fracture is nearly flat. The arkosie
material described in the section on "General Geology"
occurs under this fault.

Between, Tioineh 3 and 4. Between tunnels 3 and 4
the track was laid mostly on fill. The rails between the
north side of this fill and the south end of tunnel 3 had
been contorted into bends with radii of 20 feet or more,
as if the ground under the track had been shortened.

No fractures were observed under the tracks, but a
small normal fault was observed in the bedrock about
300 feet east of the fill between tunnels 3 and 4. The
vertical displacement on the faidt appeared to be about
2 feet. On July 22, its strike was N. 80° W. and its dip
50° SW— into the hill. On August 14 it had a similar
strike, but the dip had flattened to 37° SW, probably
by slumping.

Tunnel 4. Tunnel 4, originally 334.4 feet long, runs
about east southea.st. Its walls were cracked or broken
from the west portal to a point about 85 feet from the
east portal. Mitlway between the portals, in a zone about
50 feet wide, large breaks had occurred and the walls

Part I]

Geology

1 fau
ced

woll

■7

-Q

t^-

C/)

0 .^ 0
c -0 w

S Ot i ai sit 1

*» «

0 SS

1 rs ir
1 0 1

5 —

0 3
wo

1. -

•'T*

Ll_

0 E

Bo.../

1/ "'

^.5IB

0
0

_J
tiJ

i«

z>
1-

1
1
1

Ect

t-i

Earthquakes in Kern County, 1952

[Bull. 171

(A) Normal

(B) Tunnel 3

(D) Tunnel 5

(E) Tunnel 5

(F) Tunnel 6

Figure 3. Typical tunnel cross sections to show results of
earthqual;e damage. All sections drawn looking along track toward
Tehachapi.

the bulldozers had made the extensive excavations around
the tunnel, the fault was visible in the bedrock from the
top of the cut to track level, a vertical distance of more
than 100 feet. Fiftv feet above the track level in this
cut the fault strikes N. 67° E. and dips 72° SE.; it
is marked there by a gouge zone 1 foot to 3 feet wide
and a much wider breccia zone. On the south (hanging-
wall) side the rock is a hard and relatively unaltered
gabbro. On the north side it is highly altered and light-
colored ; the amphibole is altered to pale green epidote( ?)
and only the subordinate biotite is unaltered. The rock in
the two walls appears to be identical except for the
alteration. It seems probable that at the time of altera-
tion both walls were affected equally. As the degree of
alteration is now very unequal, previous movements on
the fault are suggested. If the alteration is assumed to
result from surface weathering, then the fresh rock must
have come from depth, and the older movements must
have been of the reverse type, and current movement on
the fault is but a renewed action along the old line of
weakness.

It\»^'^ ?»■

<»-v

»<

^-'^'

1 . >

^ 'Ml

-"r^

l<\\^

.^i

**»\* '

^^^'Lk^smsmmi^

FlGUItE 4. Air vii'w (if till' S-(Mirv<' in the Soutlicrn I'aoitir Ka)lii>:i<i iimcks rasi c.t 1'
fornia. View .south. The main valley separate.^ tunneks '.', and 4 on the extreme right side
5 and C left of center. Portal of tunnel G is distorliMl. Photo hy John Shilloii.

ealviile, Cali-
from tunnels

had buckled inward, giving the tunnel a cross section
resembling a keyhole.

The largest vertical displacement observed in any of
the tunnels occurred on a fracture about 95 feet from
the east portal of tunnel 4 ; the tioor and walls were dis-
placed about 3 feet vertically and 8 inches horizontally.
The fracture trends east, dips 72° S., and is a nor-
mal fault along which the rocks on the south side moved
relatively down and west. The track remained unsevered
and wasleft suspended in the air above the downdropped
block.

This normal fault was visible on the surface above the
tunnel where it could be traced for about 500 feet. After

Between Tunnels 4 and 5. On curve 38, south of
tunnel 4, the damage consisted of small landslides and
till shifting. On curve 39, between tunnels 4 and 5, the
track was laid on a deep fill across the main valley. After
the earthquake the track and ties were a foot and a half
south of their original cast marks in the fill. Appar-
ently the fill under the track shifted during the earth-
quake, while the arc of track, firmly tied to bedrock at
each end, did not.

Tunnel 5. Tunnel 5, which runs northwestward from
the west portal, caved in at three places and was inaces-
sible for many days after the earthquake. When visited

Part I

Geology

:^-^:

Figure 5. Damage near the east portal of tunnel 3. Apparently
the wall was raised just as the track bent, and the wall then came
down on top of the rail. View northwest.

on Augiust 14 it had been reopened enougli for a man
to walk or crawl through. The principal damage was
in a zone 833 to 974 feet from the west jiortal. The fol-
lowing notes describe the tunnel damage as observed on
August 14. The intervals (in feet) refer to postearth-
quake distances from the west portal. The right wall is
the northeast wall.

Feet

0 to 220
220 to 270
270 to 2!»0
2!)0 to 450
4M to 475
475 to 500

Minor breaks in concrete at base of tunnel walls.
No damage.

Minor breaks in concrete of left wall.
No damage.

Minor breaks in concrete walls.
Major breaks. A fault marked by a 14-inch gouge
zone crosses tunnel in this interval from right wall
(475 feet) to left wall (500 feet). It strikes about
X. G0° W. and dips 65° XE. Reinforcing bars in
the concrete were bent into a "U" indicating short-
ening of 6 inches, and the ceiling arch is 6 inches
lower in the northeast block than in the southwest
block. This is a normal fault, yet seems to have
resulted in horizontal compression.

500 to 598 Concrete cracked.

598 to 632 Roof collapsed and tunnel originally closed but now
opened.

632 to 833 Tunnel open but walls badly damaged. Center of
left wall buckled inward 3 to 5 feet. Track bowed
up from floor in a broad arch whose center is IJ
feet above grade. Maximum distortion of walls at
700 feet.

833 to 850 Roof collapsed and tunnel completely closed. Small
passageway reopened.

850 to 9(50 Very badly damaged. The arch of the roof had not
completely collapsed, so section is not closed. Right
wall, however, is pushed over against the left wall,
leaving only tall narrow opening. In a typical cross
section the lower part of the tunnel is only 1 foot
to 4 feet wide, but just under the roof arch the
opening widens to 8 feet. Normal tunnel width is
16 feet.

960 to 974 Roof collapsed and tunnel completely closed. Small
passageway reopened.

974 to 1170 No damage.

1170 East portal (actually west of other portal).

Between Tunnels 5 and 6. Between tunnels 5 and 6
the track was mostly on fill, which settled and moved
downhill slightly and carried the track with it. In the
middle of the fill zone the track both slumped and moved
laterally a foot or two.

'•^'5;

■■%.■.

^>ir^^

FlouRE 6. Sharp bend in the tracks south of
tunnel 3 strikingly demonstrates that the ground
had been shortened in this area. A landslide
blocks the east portion of tunnel 3.

Figure 7. View in tunnel 4. Normal fault
dipping 72° toward the camera has offset the
walls and lifted the track 3 feet. The slender
curved rods are steel reinforcement.

Eaethquakes in Kerx Coun'tt, 1952

[Bull. 171

^" -^C:?.

Figure 8. Surface expressum ot the normal taiiir simwii m
figure 7. The fault was traced several hundred feet to the right of
the area shown in the photograph. View north.

Tunnel 6. Tunnel 6 was not on the main fault zone.
Its walls were shattered and broken, mainly by longi-
tudinal cracks in the arch, and the tunnel was displaced
to the left with respect to the tracks. In the hillside
above the tunnel there were fractures roughly parallel
to the tunnel, suggesting that the ground through which
the tunnel passed slumped downhill a few feet as a re-
sult of the earthcjuake, while the track moved somewhat
less. Tunnel 6 has been daylighted.

Beyond Tunnel 6. Beyond tunnel 6, in the east arm
of the S-curve, the track was warped by settling of fills
and covered in places by landslides ; otherwise it was not
seriously damaged. Where the fault zone crosses the
track, the rails were slightly buckled and the fills had
slumped 2 feet. Thirteen inches of rail were removed
from the tracks during realignment. There are fractures
trending N. 53° E. on both sides of the track in this area.

As tunnel 7 lies southeast of the general area described
above and is not on the main fault zone, it was not
visited. The concrete walls of the tunnel are said to have
been cracked by the initial shock, but not broken. Later
shocks, however, worked some of the concrete loose and
necessitated repairs.

FiGlRE 9. Cracked and buckled walls in tunnel "i. Camera is on
the cave-in at 833 feet from the west portal ; looking toward the
cave-in at 632 feet. The tracks are bowed up about IJ feet.

EVIDENCE OF SHORTENING AND VERTICAL DIS-
PLACEMENT IN THE FAULT ZONE

The major fault damage in the vicinity of the South-
ern Pacific tracks near Bealville was confined to an east-
west zone about 500 feet wide. The fault zone was well
delimited where it cut tunnels 3 and 4, and also where
it cut tunnel 5, though at the surface its boundaries were
obscure.

Tunnels 3 and 4. The most spectacular evidence of
crustal shortening was that afforded b\' twisted and con-
torted rails found near the east portal of tunnel 3, where
measurements made by the Southern Pacific engineers
the day after the earthquake indicate that the earth's
crust was shortened by 10.9 feet.

Measured shortening near tunnels S and 4.

S.P.R.R. (tape) U.S.G.S. (.stadia)

Tunnel 3 2.3 feet 2.0 feet

Between tunnels 3 and 4 8.6 feet 9.6 feet

Tunnel 4 0 —1.7 feet

Total shortening 10.9 feet 9.9 feet

All this shortening occurred between the northern-
most break in tunnel 3 and the west portal of tunnel 4;
outside of this zone neither the rails nor the ties had

1000

DISTANCE IN FEET

Kioi'iiE 10. Profile along tlie tracks showing grade l)efore and after earthi|uakp. Datum is west portal of tunnel 3. Tunnel positions indi-
cated arc those immediately following earthquake. Vertical exaggeration xlO.

Part I

Geology

shifted in relation to the ground or to each other. About
75 percent of tliis shortening oecurred in a section less
than 150 feet long, just outside the east portal of tun-
nel 3. South of this section, the shifting and deformation
of the rails was minor; north of it, tunnel 3 was short-
ened only about 2 feet. In spite of the concentration of
shortening in this narrow zone, however, no significant
fractures were observed, even though half of the section
is underlain by bedrock.

The observed vertical displacement of 3 feet on tlie
normal fault in tunnel 4 should have caused an extension
of 0.6 foot along the line of the track. The Southern Pa-
cific Company' 's measurements, however, show no change
in the total length of tunnel 4, so this extension in one
part of the tunnel was apparently balanced by shorten-
ing in other parts.

Figure 10 illustrates the relative vertical displace-
ments along the track line of tunnels 3 and 4. The
dashed line represents the theoretical prefanlt grade of
2 percent, and the solid line represents the postfault
grade as determined with the plane table. The elevation
of the west portal of tunnel 3 was assumed to have been
unchanged and was used as a datum for the horizontal
and vertical measurements. Any actual change in the
elevations of these portals will be determined when the
U. S. Coast and Geodetic Survey completes its present
resurvey of tlie first-order-level line through the area
(Whitten, this bulletin).

Apart from the displacement on the normal fault at
the south end of the disturbed zone, the principal eiTeet
of the earthquake on the track profile was to warp the
track into an arch, whose crest was about 3| feet high
and coincided with the contact of the fill and the bedrock.

Tunnel 5. No complete resurvej' of tunnel 5 had been
made at the time of the authors' last visi|, but enough
work had been done by the Southern Pacific engineers
to indicate some liorizontal and vertical shifting. The
eenterline of the tunnel was displaced horizontally about
1 foot. The bent reinforcing bars at 486 feet and some
bowed-up track between 632 and 833 feet indicated local
shortening, but the overall length of the tunnel was re-
ported to be unchanged.

According to a preliminary survey, the arch of the
tunnel dropped 6 inches at one point and 3 inches at
another. In one zone the grade of the arch, which was
formerly 2 percent, had changed to 0.5 percent for 50
feet and to 4 percent for 90 feet, perhaps because of
slumping.

Other Areas. There is a marked contrast between the
observed characteristics of the fault zone in the tunnels
and on the surface. In the tunnels the zone was about 500
feet wide and the damage was severe. On the surface,
except along the normal fault over tunnel 4, the frac-
tures were few and small and the displacements were
less than a foot.

Where the fault zone crosses U. S. Highway 466, it is
about 300 feet wide, but the only visible damage con-
sisted of 5 or 10 minor fractures in the asphalt. Many
small normal faults were seen between the highway and
the railroad, and the displacement on each was less than
a foot.

In the east arm of the S-curve in the railroad, where
there are no tunnels, the faulting caused only a single
minor fracture and a 13-inch shortening of the rails.

SUMMARY AND CONCLUSIONS

In the jireceding jiages the writers have outlined and
described in detail the earthquake damage along the
Southern Pacific tracks and have presented a minimum
of interpretation. In the following section this damage
is grouped first according to type and then according to
distribution; each grouping, in turn, leads to sugges-
tions of geologic causes. In the concluding statement, all
the.se data and the deductions drawn from them are
.synthesized to produce an interpretation of the process
that caused all the observed features.

Evidence Given by Types of Damage. 1. Damage by
landslides was widespread. Tunnel 6, north of the fault
zone, was damaged by a landslide, and many stretches of
track in other areas were covered or dislocated. Some of
the wall buckling in tunnel 5 may have been caused by
the pressure of sliding earth. The prime cause of this
damage was undoubtedly the dislodging of ma.sses of
regolith and decomposed rock by the earthquake.

2. Damage clearly related to a single fracture was
found in only two places : at 239 feet in tunnel 4, and
at 475 feet in tunnel 5. Both fractures are normal faults.
In view of the evidence that the net result of motion in
the fault zone was shortening, it is believed that normal
faulting was a late and minor phase of the total move-
ment ; reverse faulting occurred first, and normal fault-
ing occurred slightly later as a result of settling.

3. The most extensive damage to tunnels consisted of
buckling and cracking walls. As this type of damage was
confined almost exclusively to the fault zone, it probably
was not caused directly by earth shaking, but rather by
displacements along a multitude of small fractures,
which may have been the distributaries of larger move-
ments at depth.

4. The most striking damage, and the most significant
tectonically, was the buckling of the track between tun-
nels 3 and 4. The ground beneath the track had not
only been shortened over 10 feet but had been bowed
up. Both of these features prove that there was at least
local compression.

Evidence Given by Distribution of Damage. 1. The
principal shortening in the vicinity of tunnels 3 a)id 4
took place in a strip 150 feet wide, located in the north-
ern part of the 500-foot-wide fault zone. In tunnel 5,
also, the major damage took place in the northern part
of the fault zone. This asymmetrical distribution of dam-
age could be explained by one of the following hypoth-
eses: (a) The stress was most intense along the north
boundary of the fault zone, and diminished southward
as if the shock had been instantaneously relieved along
numerous fractures. This hypothesis hinges on the prem-
ise that faulting had occurred in this zone previously
and that the renewed stress was relieved mainly in the
northern part of the fault zone, (b) The major shorten-
ing may have occurred in the first instant of shock over
a zone about 150 feet wide. In the next few moments
the fault zone mav have widened southward to the 500-

Earthquakes in Kern County, 1952

[Bull. 171

foot width later observed, giving the appearance of
asymmetrical destruction in the final zone.

2. The fault zone as exposed in the tunnels was about
500 feet wide, but on the surface it was apparently much
narrower. This narrowing- of the fault zone upward
probably was more apparent than real, for minor frac-
tures in rigid concrete would be seen and recorded,
whereas many of these fractures might pass unnoticed
when they cut the overlying regolith. Near the surface,
movement was dissipated along numerous fractures, and
part of it may have been absorbed by intergranular
shifting in the regolith and decomposed bedrock.

3. Large-scale shortening was found in the vicinity
of tunnels 3 and 4 but nowhere else along the strike of
the fault zone in this area. Compression may have been
concentrated in the vicinity of tunnels 3 and 4, and
have been dispersed to the east and west, because of
lateral variations in the competency of the bedrock. The

presence of fill between the two tunnels may have made
the effects of compression more apparent.

Conclusions. The earthquake that occurred in the
vicinity of Bealville on July 21, 1952, was the result of
compression that was relieved by reverse faulting in the
White Wolf fault zone. Because the south block moved
up relative to the north block (fig. 12), the fault zone
dips south. Seismologic and other data * published after
this paper was written corroborate this dip and move-
ment. The compression was immediately followed by
relaxation and settling along normal faults.

• Buwalda. J. P., and St. Amand, Pierre. 1953, Arvin-Tehachapi and
Bakersfield earthquakes of July-August 1952 : Bull. Geol. Soc.
America (abstract p. 1500, Dec. 1953) ; Richter, C. F., 1953,
Kern County aftershocks : Progress Report : Geol. Soc. America
(abstract in press) ; and Dibblee, T. W., Jr., and Oakeshott,
G. B., White Wolf fault in relation to geology of the southeast-
ern margin of San Joaquin Valley, California ; Bull. Geol. Soc.
America (abstract p. 1502-03, Dec. 1953).

8. MEASUREMENTS OF EARTH MOVEMENTS IN CALIFORNIA*

By C. a. Whitten t

Ahstracts. Resurvpys by the United States Coast and (Jeodetie
Survey across parts of the Sail Andreas fault are consistent in
showing a slow drift to the nt>rth\vest at a rate of about '2 in<'hes
per year, west of the fault. Reobservation. in 1941, of the tri-
an^ulation system crossing the San Andreas fault in Imperial
Valley, where the earthquake of 1940 occurred, established the
fact that the area on the east side of the fault shifted to the
southeast and that the area on the west shifted to the northwest.
Preliminary results of repeat surveys of triauKulation and level
schemes in Kern County, in September 19.~>S, suggest that the
Bear Mountain block, southeast of the White Wolf fault, moved
north-northeast a distance on the order of one to 2 feet and the
valley block a similar distance in a west-southwest direction. The
Bear Mountain block was also elevated, and the valley side was
depressed on the order of a foot and a half near Arvin.

The large relative displaeemeiit.s in the earth's crust
which were noted after the San Francisco earthquake
of 1906 suggested the repetition of surveys for deter-
mining the amount and extent of these horizontal move-
ments. Reports from residents indicated relative dis-
placements from 5 to 20 feet at many points along the
San Andreas fault (fig. 1). These relative displacements
were noted along 185 miles of the fault and averaged
about 10 feet.

Becau.se of the changes in geographic positions of
points near the fault, it was necessar3' for the Coast and
Geodetic Survey to reobserve the existing triangulation
in that locality. The first surveys in the area had been
made in 1851. The basic first-order scheme had been
completed in 1885. By noting the differences in the geo-
graphic positions of the triangulation stations as deter-
mined by the resurvey, it was possible to measure these
displacements.

The report of the Superintendent of the Coast and
CTCodetie Survey for 1907 contains a detailed description
of these resurveys with tabulations, maps, and sketches
showing the differences of the geographic positions as
determined by the various surveys. The studies made
in 1907 produced unexpected evidence of earlier dis-
placements, probably the result of the earthquake of
1868.

In 1922, at the request of Dr. Arthur L. Daj', director
of the Geophysical Laboratory of the Carnegie Institu-
tion, and chairman of the Committee on Seismology of
that Institution, the Coast and Geodetic Survey made
plans to reobserve the first-order triangulation scheme
along the coast between San Francisco and Los Angeles.
These resurveys were more extensive than those made
immediately after the earthquake of 1906 and were
completed in 1924. The results showed very conclusively
that there had been relatively large displacements. Be-
cause of the length of the scheme and the possible ac-
cumulation of errors, it was not possible to determine
the absolute amount and direction of the movement for
the points in the middle of the arc.

In 1929, the Committee on Seismology recommended
the establishment of a series of arcs of triangulation
crossing the San Andreas fault at right angles, with
repeat observations at 5- to 10-year intervals. Each

* Modified from "Whitten, C. A., Horizontal Earth Movement, in the
Journal of the Coast and Geodetic Survey, April 1949, no. 2, pp.
84-88. Later data relating to the Kern County earthquakes were
furnished by Mr. Whitten in September 1953.

t Mathematician, U. S. Coast and Geodetic Survey.

arc was to consist of a primary scheme, 40 to 50 miles
in length, supplemented with a secondary scheme of
closely spaced points inside and extending the full length
of the primary scheme. This pattern of survey will aid
in measuring two types of earth movement. The primary
scheme when reobserved will indicate the presence of
any movement or drift of the area on one side of the
fault relative to the area on the other side. If this move-
ment is continuous over a long period of time, the re-
peated observations will show the rate of movement. The
repetition of the survey with the secondary scheme of
closely spaced points will measure these smaller move-
ments.

The first two of these special surveys were established
from Newport Beach to Bear Lake (1) and from Point
Reyes to Petaluma (2). This work was completed in
1929 and 1930. During the next 3 years four similar
projects were extended from Monterey Bay to Jlari-
posa Peak (3), from San Fernando to Bakersfield (4), in
the vicinity of the San Luis Obispo (5), and in the
vicinity of Taft (6).** In 1934 the scheme between New-
port Beach and Bear Lake was reobserved. An investi-
gation of these reobservations made at that time indi-
cated there had been no displacement of any significance.

The are of triangulation between Point Reyes and
Petaluma was reobserved in 1938. The results of this re-
survey were not conclusive, although some interpreta-
tions gave evidence of a northwesterly drift for the
stations in the vicinity of Point Reyes.

The Committee on Seismology made further recom-
mendations in 1935 to modify the pattern of the sur-
veys. The new plans specified lines of traverse and level-
ing crossing the fault at right angles. The marks were
to be spaced at intervals of 100 feet for the first mile
from each side of the fault, at 200 feet for the second
mile, and at 300, 400, and 500 feet for the third, fourth,
and fifth miles, respectively. Eight areas located along
the San Andreas fault in southern California were
selected for these special studies. The areas were near
Maricopa (7), Gorman (8), Palmdale (9), Inglewood
(10), Brea (11), Cajon Pass (12), Moreno (13), and
White Water (14). The surveys for the Maricopa, Gor-
man, and Palmdale zones were completed in 1938,

The traverse in the vicinity of Palmdale was remeas-
ured in 1947. A comparison of these measurements with
those of 1938 disclosed no changes indicative of earth
movement. The small differences which were noted were
either the result of accidental errors of observation or
due to local settlement of marks.

The surveys in the vicinity of Maricopa and Gorman
were reobserved in 1948-49. It is planned to continue
this project of establishing and repeating these special
traverses until the eight zones are completed.

Imperial Valley Earthquake. On May 18, 1940, a
severe earthquake occurred in the Imperial Valley. Al-
though no triangulation had been established in that
area for the particular purpose of studying earth move-
ments, an extensive net covering the area had been

•• Note: The numbers in parentheses refer to the areas correspond-
ingly numbered in fig. 1.

( 75)

Earthquakes in Kern County, 1952

[Bull. 171

-a-

-IT

-as-

-34*

-sr

1 1

i2r ur

+ »•-

+ !!■-

+ ir-
0

Poinl R«yes

^frCi]

V)Sn FRANCISCO

+
+
+
+

\(3I Y^O^X "^

-9

>
+

XASan Luis 0

pa. "

119* lir 117* US* Hi* 36*-
T -t -t -^

ANDREAS FAULT

"" ^T^

W^^

o ^ ^\/\\

^^^^-;^T\ t|^ Son Fernando K" ^^^

+ ^^v\3

123*

SCALE OF MILES

0 10 20 30 40 50 60 70 80 90 lOOMILES

ISL

120*

AND '"*'**^r^^n?l^^^^_13^°^"'w^i \

w m-

-" c K 1 C 0\

M!- n«- \ iir M t " lis-

1 1 ._ i L 1 1 1

Figure 1. Hachured areas show iilaniicd control across San Andreas fault to determine crustal changes. Xumerals
in parentheses refer to corresponding numl)ers in text. Reprinted from C. S. Coast i£- Geodetic Survei/ Journal, April
10.',!), p. 8-',.

Part I]

Geology

5 4 3 2 10

STATUTE MILES
5

SCALE OF VECTORS IN FEET
1 0 1 2 3 4 5 6 7 8 9 10

FAULT LINE

OFFSET 225

Figure 2. Earthquake investigation in Imperial Valley, California, 1941. Reprinted from V. S.
Coast & Geodetic Survey Journal, April 19^9, p. 86.

Earthquakes in Kern County, 1952

[Bull. 171

SCALE OF VECTORS IN fEH

SCALE OF IRIANGULAIION IN FEET

5O00O 100000 ISO

Figure 3. Results obtained from four triangulation adjustments
between 1882 and 1046 showing displacements as vectors. Ueprinted
from v. 8. Coast d Geodetic Survey Journal, April 19^9, p. 87.

established in 1935, with supplemental surveys in 1939.
A part of this triangulation was reobserved in the spring;
of 1941. After the work was completed, a preliminary
investigation indicated that the resurveys should have
been extended over a larger area so that the problems
of adjustment would be simplified. No further field work,
however, was done at that time.

At the conclusion of the war the data from the sur-
veys in this area were given further .study. Comparisons
of the final geographic positions of the two adjustments
sharply defined the location of the fault line, the direc-
tion and magnitude of the horizontal movement, and
the extent of the area that was affected by the.se move-
ments. The vectors in figure 2 show the direction and
magnitude of the displacements. The region of maximum
shift was near the southern limit of the survey. Reports
from Mexico stated that the amount of displacement de-
creased along the fault south toward the Gulf of Cali-
fornia. It can be seen from the figure that the area on
the east side of the fault shifted to the southeast and
that the area on the west shifted to the northwest. At
distances of 15 to 20 miles east or west of the fault the
magnitude of the shifts is reduced to a small fraction of
a foot. The data from this investigation are more con-
sistent in showing these displacements than are the re-
sults of any previous resurvey. Tliis study brought out

the great value in having an extensive triangulation
network over all of the area of the fault, so that if an
earthquake did occur, the basic surveys will have been
made.

Slow Drift Along Coast. In 1946 basic triangulation
networks were executed in the San Francisco Bay area
and in the Santa Clara Valley with rigid connections
made to the old primary scheme. A comparison ef the
lists of the directions from the various surveys spaced
over a period of more than 60 years shows that there
has been a progressive change in the azimuths of the
lines crossing the faidt at approximately right angles.
The azimuths are increasing in a clockwise direction.
Astronomic azimuths observed in 1885, 1906, 1923, and
1947 on one of the lines crossing the fault also show
this progressive change. Since azimuths determined by
triangulation and those determined astronomically are
independent of each other with regard to observation
and computation, the similarity of results strengthens the
evidence supporting a slow drift of the area to the west
of the fault. Knowing the lengths of the lines crossing
the fault, the displacements needed to produce the
changes in azimnth were computed. The results of these
computations are very consistent and show a slow drift
to the northwest at a rate of about 2 inches per year,
west of the San Andreas fault. The width of the area
varies from 30 to 40 miles. This rate is based on the
results of the four different surveys spaced at intervals
of approximately 20 years. (See fig. 3.)

The results of these studies showed the need for more
extensive surveys so tliat the geographical limits of the
areas affected by this movement could be determined.
In 1948 the triangulation scheme north of San Fran-
cisco was reobserved as well as the scheme extending
along the coast as far south as San Luis Obispo and then
east to Bakersfield. The same slow movement is indi-
cated throughout the ftdl length of the scheme. The
section near Bakersfield was originally observed in 1926.
The other surveys which have been repeated date back
to 1880. The longer span of years of course will give
a more accurate rate. However, even the more recent
work near Bakersfield shows a rate for this movement
of an inch and a half per year (the area east of San
Andreas fault is moving south). It will be necessary to
repeat these survej^s after an interval of a few years to
verify this rate.

Kern County Earthqnakes of 1952. After the Arvin-
Tehachapi earthquake of July 21, 1952 and numerous
aftershocks, including the Bakersfield earthquake of
August 22, the IT. S. Coast and Cfcodetic Survey made
repeat surveys of triangidation and level schemes in
Kern County. Field work was started in October 1952
and completed in January 1953 ; adjustments of the
surveys were in progress in September 1953 when this
section was written. Some of the final results of the tri-
angulation adjustment and preliminary results from the
releveling are shown graphically in figure 4. The line
marked "White Wolf fault" is the trace of the fault
based on geological field evidence and has been added
to the map by tlie Division of Mines. The horizontal dis-
placement, as determined from adjustment of the 1951-52
and 1952-53 surveys, is shown by means of vectors. The

Part I]

Geology

Earthquakes in Kekn County, 1952

[Bull. 171

vectors mitrht be expected to have errors equivalent to
half a foot or possibly a foot. The relationship between
two adjacent points where the shifts are shown by the
vectors may be considered to be accurate to a quarter
of a foot. The triangulation stations on Double Moun-
tain and on the high point about 2 miles north of
Tehachapi Pa.ss were used as fixed or stationary points
in the adjustment.

The differences of elevation are the result of the com-
parison of the two surveys without an adjustment of
closures. The sharp break about 6 miles south of Arvin
has been accurately determined as well as the more grad-
ual uplift southwest of that point. The area of subsidence
south of Bakersfield is not definitely determined with
respect to its geographical extent, but the magnitude of
the settlement is accurate to a small fraction of a foot.
The vertical changes occurring through the mountains
between Bakersfield and Tehachapi are not as sharply
defined.

As may be seen in figure 4, the data show that the
Bear Mountain block, southeast of the fault, moved
toward the north-northeast a distance on the order of
one to two feet, but the southwest segment of that block,
as shown by triangulation stations, appears to have
moved upward and toward the northwest over the valley.
The one triangulation station on the valley floor suggests
movement of the valley block a similar distance in a
west-southwest direction.

Greatest vertical movement, an elevation of 2 feet, ap-
pears to have taken place in the epicentral region of the
Arvin-Tehachapi earthquake of July 21, with the Bear
Mountain block elevated and tilted toward the south-
east, but moved northwest. Depression of the valley side
was on the order of a foot and a half, centering in a
basin-shaped area southwest of Arvin.*

• Ed. Note. It is significant that these measured movements of the
land surface, showing the southeast (Bear Mountain) block
moved relatively up and in a northerly direction, are thoroughly
consistent with geologic data indicating the White Wolf as a
left lateral reverse fault, and with seismologic data supporting
oblique-slip movement in the same sense on that fault.

9. EFFECT OF ARVIN-TEHACHAPI EARTHQUAKE ON SPRING AND STREAM FLOW

By Revoe C. Briggs t and Harold C Troxell t

ABSTRACT

Flow in ninny of the streams and springs in the urea covered
by this report inireased as a resnlt of the Arvin-Tehachapi earth-
quake. Although tliis increase in flow appears to have been tem-
porary, there was still evidence of it in some of the streams and
springs as late as June \i)'>H, when this reimrt was prepared. It
is doubtful if the earthiiuake had any permanent effect on the
recharRe areas or on the permeability of the aquifers. This tempo-
rary increase in some cases is probably due to the mere disturbance
of the nncon,soli(late<l material in the di.scharKe areas, resultiuK
in the clearing of the existing outlets and opening of new ones.

INTRODUCTION

111 arid and seiuiarid localities where water supply is
always in the public mind, any event which affects nat-
ural water resources is of interest. The Arvin-Tehachapi
earthquake proved to be no exception. Immediately after
the earthquake local newspapers reported marked in-
crease or decrease in the flow of several springs and
streams.

From time to time data on the flow of springs affected
by earthquakes have been obtained by private and pub-
lic agencies. Usuall.v these individuals or agencies do not
have an opportunity for placing such data in public rec-
ords and consequently much information is unavailable
for use by the general public.

This analysis represents an attempt at reporting all
data available at this time reflecting the change in flow
of the mountain streams and springs in Kern, Santa
Barbara, Ventura and Los Angeles Counties as a result
of the Arvin-Tehachapi earthquake. However, neither
the degree of coverage nor the type of data is identical
throughout this entire area.

A special effort was made to obtain factual data in
Kern County near areas of greatest disturbance. How-
ever, the available data consisted primarily of eye-wit-
ness accounts of changes in spring or stream flow. These
accounts were supported by continuous records of dis-
charge for a number of streams at gaging stations main-
tained by the U. S. Geological Survey or the U. S.
Bureau of Reclamation. The points of observation and
reference numbers for data in Kern County are shown
on figure 1. In addition, this map indicates by symbol
whetlier the flow of the spring or stream (a) increased,
(b) decreased, or (c) remained unchanged, as a result
of the Arvin-Tehachapi earthquake.

By far the predominant effect of the earthquake on
these streams and springs in eastern Kern County was
to increase the flow. Probably the most noteworthy evi-
dence of this increase was in Caliente Creek basin. Be-
fore the earthquake the stream channel of Caliente
Creek (25) was completely dry below its confluence
with Tehachapi Creek at the town of Caliente. Immedi-
ately after the earthquake, the flow from springs in the
headwaters increased so that within a few days the flow
of Caliente Creek at Caliente reached about 25 cubic
feet per second and remained near that value until win-
ter precipitation increased and sustained a still larger
flow.

• Published by permission of the Director, U. S. Geological Survey,
t District Engineer and Hydraulic Engineer, U. S. Geological Survey.

Ill Santa Barbara, Ventura, and Los Angeles Coun-
ties, it was iiiqiractical to engage in a complete field ex-
amination. As a result this study was restricted to the
mountain areas of these counties. The data used were
obtained from a well distributed network of gaging sta-
tions operated cooperatively by the U. S. Geological Sur-
vey and the State of California, suiipleineiitcd by sta-
tions operated by Los Angeles County Flood Control Dis-
trict and Ventura County Water Survey. This network
was further supplemented by an intense investigation of
stream and .sjjring flow already under way in one area.
In the Santa Ynez Mountains just west of the city of
Santa Barbara, the U. S. Geological Survey, in coopera-
tion with the Santa Barbara County Water Agency,
measures monthly the flow in about 130 springs and
small mountain streams. A portion of these networks is
shown on figure 2. The sites indicated on figure 2 include
all the gaging stations at which there was a measurable
increase in discharge attributable to the Arvin-Tehach-
api earthcpiake. Also shown is a limited group of stations
at which there was no measurable increase in flow attrib-
utable to the earthquake. Inclusion of this latter group
of stations was largely for the purpose of delineating or
defining the general area in which stream or spring flow
was affected by the earthquake.

In the mountain areas of these three counties, the
most significant increase in runoff attributable to the
Arvin-Tehachapi earthquake occurred in Ventura
County. Ill the 254-s(iuare-mile mountain drainage area
of Sespe Creek near Fillmore (90), the daily discharge
increased from 17 cubic feet per second on July 20 and
21, 1952, to 37 cubic feet per second on July 31, an in-
crease which apparently was entirely a result of the
earthquake. This increase in runoff amounted to 2,160
acre-feet between July 21 and September 30, 1952, and
was equivalent to 61 percent of the entire runoff for the
preceding dry water year of 1951.

The change in flow in many of the smaller springs
was even more spectacular. In one of the smaller springs
(67) on the Juan Y Lolita Raneho in the Santa Ynez
Mountains, the flow following the earthquake was about
three or four times as great as that during the fairly
wet water year of 1952.

Notwithstanding this noteworthy increase in flow,
almost 88 percent of the points of observation in the
Santa Ynez Mountains indicated no change in flow as a
result of the earthquake. Moreover, even in short dis-
tances there were often radical differences in flow char-
acteristics attributed to the earthquake.

The principal mountain ranges and major fault sys-
tems have been indicated on figure 2 as a suggestion of
geologic structure. Those gaging stations showing an
increase in flow are mostly located in the Santa Ynez
Mountains or minor ranges between the San Rafael
Mountains on the west and the San Gabriel Mountains
on the east. However, had coverage been as intense in
the San Rafael and San Gabriel Mountains as in the
middle Santa Ynez Mountains, evidence of increased
runoff attributable to the Arvin-Tehachapi earthquake
maj^ have been more widespread.

(81)

Earthquakes ix Kern County, 1952

[Bull. ] 71

KERN COUNTY LINE

► NEW ^ERNV^LLEL

SjuthT"" — -fA^

C^T'57

DDFISH
55

T<^"

SCALE

-9- Flow increased, dwe to Arvin -TchdchapL eart^ouake.
O Floix.not infli/enced by Ar-i/in-TehacKapL earth<judke.
3 Floic decreased due to Arv'in -TchachspL earthjuake .

Base from AMS Bakersficid Nl II

Figure 1. Location of selected gaging stations in Kern County.

Part I]

Geology

s
m

a
£

Earthquakes in Kern County, 1952

[Bull. 171

Streams investigated in Los Angeles County did not
show any noticeable increase in flow following the earth-
quake.

ACKNOWLEDGMENTS

The authors wish to thank their fellow workers in the
U. S. Geological Surve.y for suggestions and assistance
in preparing this report, particularly T. A. Cooper,
field engineer at Visalia, who obtained many useful data.
Also they wish to thank the many local people in the
earthquake area who furnished information, especially
Leroy Rankin of Walker Basin, C. W. Poole of Lor-
raine, and Brad Krauter of Tehaehapi.

Although it is impracticable to name all who gave
assistance, the following people should be mentioned :
W. M. Jaekle, Eng. Dept., Southern Pacific Railroad,
San Francisco ; J. G. Sinclair, Eng. Dept., Southern
Pacific Railroad, San Francisco; H. Cole, Southern Pa-
cific section foreman, Bena ; Roy Ballard, LT. S. Soil
Conservation Service, Tehaehapi ; Elmer Lyne, U. S.
Soil Conservation Service, Tehaehapi; L. E. Williams,
Caliente ; A. F. Neumarkel, near Arvin ; Lawrence
Brown, Caliente; W. T. Blackburn, Wriglev Ranch,
Tehaehapi; F. W. Nighbert, Bakersfield (White Wolf
Ranch); Jack Shepard, Humble Oil Co., Bakersfield;
Joe Prowell, Kern Rock Co., near Bakersfield ; Frank
Lawrence, Bodfish. The offices of the U. S. Bureau of
Reclamation at Fresno and Visalia furnished records of
discharge of several streams in the Keru County area.

CHANGES IN SPRING AND STREAM FLOW ATTRIBUT-
ABLE TO THE EARTHQUAKE

Tlie Arvin-Tehachapi earthquake is known to have
affected the flow in many springs and streams over a
sizeable area in the counties of Kern, Santa Barbara,
and Ventura. The earthquake, on the morning of July
21, 1952, occurred in that part of the year characterized
by minimum runoff and precipitation. During these
warm summer months, practically all flow from springs
and runoff in mountain canyons has its immediate origin
in perennial ground-water storage.

This storage is periodically recharged by the excessive
precipitation during the wetter years. For example, in
the San Bernardino Mountain drainage area of Mill
Creek, as a result of a very wet 1921-22 winter, there
was a large recharge to mountain ground-water storage
equivalent to 21 inches over the entire area. The next
measurable replenishment to ground-water storage did
not occur until 5 years later when a recharge equivalent
to 9.2 inches over the area resulted from the 1925-26
winter precipitation. As a result of these periodic size-
able contributions, seepage has been sustained, in the
form of perennial flow, even during such extended
droughts as those of 1893-1904, 1923-34, and 1944-51.

Earth tremors such as those experienced on Jul.y 21,
1952 may cause some change in the ability of these
fractured moimtain blocks to store ground water, and
frequently disturb existing conditions of permeability
in the discharge areas. For man.y years those closely
associated with water supply have known that seepage
from these mountain ground-water sources has been dis-
turbed at times by earthquakes. However, seldom is this
change in flow sufficientlj- documented to appear in
scientific literature. Consequently, it is the primary

purpose of this report to record all available data on
the change in flow of springs and streams in the nearby
mountain areas as a result of the Arvin-Tehachapi
earthquake.

These factual data are presented in the form of a
brief description of the point of observation and a state-
ment indicating the effect of the earthciuake on seepage
from springs and flow in streams. These data are given
in numerical sequence, segregated into two parts, (a)
Kern County area, and (b) Santa Barbai-a, Ventura,
and Los Angeles County areas. This segregation is
largely associated with the differences in types of availa-
ble basic data.

General Interpreation of Spring and Stream Flow Data

Before analyzing records of sj^ring and stream flow
it is necessary to develop criteria whereby effects of the
Arvin-Tehachapi earthquake can be identified. Discharge
of springs and streams in these California areas can be
extremely variable. This variability is of two distinct
types and is the result of a complicated interrelationship
between many of the physiographic and hydrologic fac-
tors.

First, there is the general annual cyclic-like pattern
of rtinoff distribution. Maximum discharge usually oc-
curs during the winter rainy season, due to the larger
rates of rainfall, the recharge to mountain ground-water
storage with innnediate seepage therefrom, and earh'
snowmelt. With the conclusion of the runoff resulting
from the rainy season, discharge tends to follow a fairlj-
well-defined recession during the warm, dry summer
months, with minimum discharge occurring in late sum-
mer or early fall. This recession in discharge seldom
changes in pattern and can usually be readily established
after a number of years of observation.

The second type of variability in discharge is far more
difficult to evaluate. This fluctuation varies from hour
to hour and day to day, reflecting changes in the evapo-
transpiration opportunity and the occasional summer
precipitation. This summer precipitation is generally
light and the runoff attributable to it is minor and read-
ily definable. The effect of the more significant evapo-
transpiration opportunity on flow of springs and streams
is harder to recognize. The influence of these water
losses is more generally recognized in streams or springs
where the discharge is small. Under these conditions,
changes in evapotranspiration opportunity may cause
tremendous percentage differences in day-to-day flo\v.

Consequently, certain data have been assembled on
figure 3 to illustrate the eft'ect of some of these influences
on spring and stream flow. The upper portion of this
diagram indicates days on which daily precipitation ex-
ceeded 0.1 inch at a group of mountain, or near-moun-
tain, stations for the period of June through September
1952. These records indicate the storms to be of the
usual summer convectional type and to be of limited
distribution, and are furthermore confined to the latter
part of July, August and September.

The second part of figure 3 shows daily evaporation
at the standard Weather Bureau station on the Backus
Ranch near the town of Mojave. In addition to indicat-
ing evaporation from a water surface, this record be-
comes an index of the amount of water required to
support native and domestic plant life. Variability of

Part I]

Station
Kern River P. H. 3
Kernville
Tetiochapi
Mojave
Tejon Rancho
Pattlway
Cuyama
Ozena
Ojoi

Sandberg
Mt. Wilson
Squirrel Inn No. 2

GB:oLOGy
Prec

cipitQtion

i 11 o

I 1,

1 1- "

1 ,1

; ir ^

1 1 o-

■

Explanation
Q Daily precipifotion
of on inch, or more

1 —

o oo 1

Evaporation

i -4

a. 6

1 .

k BacKus Rai

ICtl

f '

-k-L

y-\^

\ A

1 /

kJ^

1 1 '

'IV

1 N

)

1 ; '

100

lOff

1 .

i 1 , !

1 D

1 ' 1 ■

V.,V.Vi ^

LlJ

1 : ; !

40
30

•-^

■ 1 . l-N—

^, ^

o20

"^ ■ ^

^ 1

1 1

V. 15

-*\

[•Decrease in discharge due to

' — ^ — \ —

Sio

->''" "^

' incr
1

josed evapotranspiration losses

vTVx^

^ R

. , — i

•' A

_K- increase m _
! discharge du
1 to precipitat

e
on

, '1

1 1

1
1

; 1

1
1

1 /

•

-H

U-ln

crease in

disch
ed 1
ation

arge

1
1

• due to reduc
1 evapotranspir
J losses

1.5

1.0

1
1

\i-

1.0

100
80

60
50

20
15

10
8

8
5
4

- 3

1.5

10 15 20 25
June

10 15 20 25
July

1952

10 15 20 25
August

1.0

10 15 20 25 30
September

FiGUBE 3. Precipitation, evaporation, and runoff at selected stations.

Earthquakes in Kern County, 1952

[Bull. 171

this water reqnifement is indicated by the fact that daily
evaporation at this station ranged from 0.18 to 0.91 inch
with a daily mean of 0.48 inch durintr the 4-month period
of June through September 1952. This record is included
on figure 3 as an index of changes in the evapotranspira-
tion opportunity during this 4-month period. This record
is shown in an inverted form in order to indicate paral-
lelism with changes in discharge.

The lower part of this diagram gives the record of
daily discharge for two gaging stations along the San
Andreas fault zone in the San Gabriel and San Bernar-
dino Mountains, presumably on stream flow outside the
area of influence of the Arvin-Tehachapi earthquake.
The runoff from the first of these drainage areas, Rocl<
Creek (96) on the north side of the mountain range,
shows the typical summer recession, and the hydrograph
of this flow when plotted on seniilogarithmic projection,
such as used in this analysis, tends to approach a straight
line. In this rugged mountain drainage area the mantle
rock was sufficiently absorptive and retentive to retain all
the limited summer precipitation and also to subdue the
effects of daily changes in evapotrauspiration.

The second .stream, City Creek, is located on the south-
ern face of the San Bernardino Mountains near San Ber-
nardino. In this instance daily runoff fails to reflect the
smooth recession of Rock Creek. Here, due to a less ab-
sorptive and retentive mantle rock, the recession hydro-
graph is quite irregular, responding to rainfall during
each period of summer precipitation and reflecting sig-
nificant changes in evapotrauspiration opportunity.

It will be noted on figure 3 that on the date of the
Arvin-Tehachapi earthquake there was no immediate
change in discharge at either station. However, less than
ten days after the earthquake, on July 30, daily dis-
charge of City Creek increased to 8.5 cubic feet per sec-
ond from 2.7 cubic feet per second on July 21. The
records in the upper part of figure 3 indicate that this
increase in discharge was due primarily to iJrecipitation
and secondly to a reduction in evapotrauspiration op-
portunity. During the same period, changes in discharge
from these two causes are verj' much more subdued in
the Rock Creek drainage area.

Shortly after this storm period, daily discharge of
City Creek increased from 2 to 3 cubic feet per second
during a period of no precipitation. In this instance the
increase in discharge must have been due to a reduction
of evapotrauspiration opportunity as suggested by tlie
record obtained at the Backus Ranch. Again the influ-
ence of this change in evapotrauspiration opportunity
was not reflected in the discharge of Rock Creek.

On an earlier occasion, daily discharge of City Creek
decreased from 9 cubic feet per second on June 17 to 7.3
cubic feet per second on June 19. This decline in dis-
charge is largely attributable to increase in evapotraus-
piration opportunity as suggested by the evaporation
period at Backus Ranch.

In view of this variability of discharge during the
summer recession period in some streams, it is often diffi-
cult to properly accredit changes in flow to a single
event, such as the Arvin-Tehachapi earthquake.

Hydrographs of daily discharge have been included
as part of the analysis, when the data warrant it, in
order to show better the effects of the Arvin-Tehachapi

earthquake on spring and stream flow. In general these
hydrographs consist of two parts. The upper portion of
the diagram gives the hydrograph of daily discharge
plotted on a semilogarithmic projection similar to that
shown on figure 3. The advantages of this type of pro-
jection are that the hydrograph of the summer recession
tends to approach a straight line, and secondly, small
changes in discharge can be readily identified where these
changes are large in percentage of discharge. When avail-
able, typical antecedent records are included in order to
develop the trend of the 1952 runoff had the earthquake
not occurred.

The lower part of most diagrams shows the change in
flow attributable to the earthquake. This portion of the
diagram is plotted on an arithmetical scale, so that the
change in discharge and its time distribution is more
readily discernible. Then for purpose of emphasis, in-
crease in discharge due to the Arvin-Tehachapi earth-
quake has been cross-hatched on both parts of the
diagram.

Passing mention should be made of a seismograph
t^ype of record nearly always found on the charts of re-
cording stream-gages in an area affected by earthquakes.
At the time of each large shock a vertical line is found
on the gage-height chart, and there is a rough relation
between the severity of the shock and the length of the
line.

The line is caused by the vertical movement of the
float in the stilling-well when the shock sloshes the water
from side to side. Vertical movements indicating surges
of more than a foot have been recorded.

Following the shock, if the gage-height graph con-
tinues at the same stage as before the earthquake, the
event has no immediate significance so far as the dis-
charge of the stream is concerned.

Occasionally the change in the discharge at the gaging
station might be the result of the making or breaking of
a small dam in the stream channel iipstream from the

recording gage.

Kern County Area

In this mountain area where the Arvin-Tehachapi
earthquake damage was most significant, the effect on
flow in springs and streams was most prompt and pro-
nounced. Good evidence is available to show that the
flow appreciably increased in some 15 streams and 32
springs within about 35 miles of Caliente as shown on
figure 1. This same map indicates eight springs where the
flow is known to have decreased.

Previous mention has been made of summer precipita-
tion and the fact that generally its effect could be sep-
arated from that of evapotrauspiration and from the
effect of the earthquake. So far as eastern Kern County
is concerned, it seems quite certain that precipitation in
late July and early August was almost insignificant in
its effect upon spring and stream flow. Records of the
1 1. S. Weather Bureau, as well as testimony of local resi-
dents, showed that scattered thunderstorms visited some
parts of eastern Kern and Tulare Counties during the
period July 24-31. However, such moisture was generally
small in quantity- and fell on relatively small, widely
separated areas. Table 1 shows daily precipitation for
the period July 23 to August 2 at eight representative
precipitation stations in eastern Kern and Tulare
Counties.

Part I]

Geology

Table 1

Precipitation station

Tejon Rancho

Teliaeliapi

Wehion*

Southern California Edison Co. Kern River No. 3 Powerhouse-
California Hot Springs*

Spring\ille, Tule Headworks dam*

Cilenville Fulton Ranger Station*

Lorraine*

Daily precipitation in inches — 1952

July

0
0

(}

.03
.01

0
.00

0
.43

0
.01

T
.03

.01
.12
.30

0
T

0
0
0

.15
0
0

Augu.st

' Recording gage.

Detailed topograpliy of eastern Kern County is shown
on the topographic maps of the U. S. Geological Survey.
The respective maps are named in the following text
which gives a brief description of each spring and
stream. The numbers following the names of springs and
streams are those used on figure 1.

J'leito Creek at Snn h'liiigilio Kaiich (1). Pleito Creek drains
the north side of Wheeler Ridge, at the southernmost part of the
San Joaquin Valley, about 20 miles south of Bakersfield (Buena
Vista Lake quadrangle).

Jack Shepard of the Hunilile Oil Company, Bakersfield, stated
that Pleito Creek was dammed by a slide which resulted from the
earthquake, and for about 2 weeks thereafter did not flow at all
in the lower reaohes. The creek normally goes dry in summer and
had almost ceased flowing on July 21, 10.^>2. The earlh(iuake ap-
parently caused flow tO' start upstream from the slide-dam because
water appeared downstream from the dam about August 4 and
continued to flow throughout the summer and winter.

Grapevine Creek above Richfield Piimpintj Utatioii (2). Grape-
vine Creek drains the north and west slopes of the Tehachapi
Mountains at the extreme southern end of the San Joaquin Valley
(Tejon quadrangle). Prior to the Arvin-Tehachapi earthquake, the
U. S. Bureau of Reclamation established a temporary gaKing sta-
tion on this creek above the RichHeld pumping station. A hydro-
graph of daily discharge at this site for the period July through
, September 1952 is given on figure 4.

10 15 20 25
Jul,

10 15 20 25 31
August

10 15 20 25 30
September ,

.'- Ificreose in discharge due to Arvin
I Tehacliapi earthquake

Figure

Ilydrographs of daily discharge for Tunis Creek.

£ 2 -

•1.5 ■

O
3

/KA:

-^j

V/'i

(^ ^

^7\

-j\

10 15 20 25 31 5 10 15 20 25 30

August Septetnber

Figure 4. Hydrographs of daily discharge for Grapevine Creek.

This hydrograph shows two characteristic periods of high dis-
charge centered around July 29 and August 11. These maxima are
probably associated with reduced evapotranspiration opportunity,
and possibly a little precipitation, as indicated for City Creek. Con-
sequently it becomes difficult to deflne accurately any change in
flow in this stream due to the Arvin-Tehachapi earthquake.

The second grai>h on figure 4 represents an estimate of the in-
crease in discharge due to the Ar^■in-Tehachapi earthiiuake. This
increase in flow, credited to the earthquake, has been indicated on
the upper graph also. The increa.se in flow has been emjihasized by
cross-hatching both diagrams.

As indicated on figure 4, this estimated increa.se in flow, July 21
to September 'M), 19.52, due to the earthqinike, amounts to about
48 acre-feet, or the quantity of water reipiired to cover 48 acres
a foot deep.

Pastoria Creek above Cable Corral. El Tejon Rancho (3). Pas-
toria Creek is also at the southern tip of the San .Toa<|uin Valley,
a few miles east of Grapevine Creek (Tejon quadrangle). A
record of the flow in this stream was obtained by the U. S. Bureau
of Reclamation on the El Tejon Rancho above Cable Corral, and
indicates that discharge increased from less than 1 cubic foot per
second just prior to the earthquake to about 7 cubic feet per second
by July 26. The flow then gradually decreased to the pre-earth-
quake flow of October. The increased runoff July 22 to September
30, as a result of the earthquake, is estimated at about 200 acre-
feet.

Tunis Creek above El Tejon Rancho Diversion (4). Also drain-
ing the northwest slope of the Tehachapi Mountains, and north-
eastward of Pastoria Creek, is the Tunis Creek drainage area

Earthquakes ix Kern County, 1952

[Bull. 171

(Tejon quadrangle). The runoff from this creek above El Tejon
Rancho diversion, has been recentl.v measured l)y the U. S. Bureau
of Reclamation. A record of dail.v discharge is given on figure 5
in the form of a hydrograph. This hydrograph shows in a most
spectacular manner the increase in How due to the Arvin-Tehach-
aL)i earthquake. Within a few days after the earthquake flow of
the stream increased from 2.0 cubic feet per second to 4.8 cubic
feet per second and then maintained the normal recession slope
throughout the remainder of the period.

The lower diagram on figure 5 indicates the estimated increase
in runoff attributable to the earthquake. The cross-hatched areas
show that the increase amounted to 310 acre-feet, July 22 to Sep-
tember 30, 1952.

El Paso Creek above El Tejon Rniiclio Jlead/juarters (5). The
El Paso Creek drainage area is north of the Tunis Creek drainage
area in the Tehachapi Mountains and tributary to the south San
.Joaquin Valley (Caliente quadrangle). Runoff from that portion
of the drainage area above the El Tejon Rancho Headquarters has
been measured recently by the U. S. Bureau of Reclamation. The
daily discharge of this stream for the period of July through Sep-
tember 1952 is given in hydrngraphic form on figure 6. The effect
of the Arvin-Tehachapi earthquake on the flow was very similar

5 10 15 20 25 30
September

/^W^^77---J ^Increose in discharge due to
///////// //llJJjTy^C.,^^ Arvin~Tehachapi earthquoke

^^^^^^^^^» '

^^^^^^^^^mM'

Figure G. Hydrographs of daily discharge for El Paso Creek.

to that shown on figure 5 for Tunis Creek. The increase in runoff
attriliutable to the earthquake is believed to be well defined. As B
result of the Arvin-Teliachapi earthquake the runoff of this stream
was increased by about 800 acre-feet for the period July 22 to
September SO, 19.')2.

Baiuliii-ti Haiich Spring (6). At the Banduoci Ranch, a mile
and three-quarters southwest of Cummings Valley School (Cali-
ente quadrangle), a dry well-hole started to flow about 3.5 gallons
per minutes about August 5, 10.52. and was still flowing approxi-
mately that amount on February 17. 19.53.

Cumminpn Creek at ('iimtnin</s \'alley (7). Klmer Lyne, engi-
neer of the U. S. Soil Con.servation Service at Tehachapi, observed
Cummings Creek (Caliente quadrangle) in September, 1952. and
estimated that its flow was about 30 percent greater than normal
for that time of year.

Iitstitiitinn Spring in Cutnniingn VaUeij (S). The spring which
was the main source of water suiqily for the California Institution
for Women, on the east side of Cummings \'alley (Caliente quad-
rangle) decrea.se<l greatly in flow following the (piake. The spring
was located near the top of the ridge lietween (^'unimings Valley
and Brite Valley.

Spring in lirite Valley (9). Alongside the roail at the west
edge of Brite Valley (Caliente (piadrangle) a new spring broke
out shortly after the quake. This probalily is tied in with the de-
crease in flow, at the same time, of the sjiring which had previ-
ously been the main water supply for the California Institution
for Women, just over the hill to the southwest.

Spring in Cummings Valley (10). Several persons reported an
increase in flow of this spring at the upper end of Cummings
Valley (Caliente quadrangle).

Sycamore Canyon Creek, West Side of Bear Mountain near Ar-
rin (11). David J. Leeds. Geophysicist. V. S. Coast and Geodetic
Survey, noted an increa.se in the flow of Sycamore Creek (Caliente
quadrangle) and the same fact was reported by employees of the
Albert Angus Ranch near Arvin where S.vcamore Creek water is
used for irrigation.

Meadowbrook Farm Spring (12). At the Jleadowbrook Farm
(dairy), 2 miles west of Tehachapi (Mojave quadrangle), springs
dried up following the earthquake and sub-irrigated alfalfa died.

Wrigley Ranch Springs (13, 14, 15). At Wrigley Ranch (for-
merly Hall Ranch) 1 mile north of Old Town (Jlojave quad-
rangle), manager W. T. Blackburn reported that their large spring
had decreased over the years until it was just a trickle prior to
the quake. Almost immediately afterwards it increased to "fill a
4-inch pipe," and two new springs appeared north of the original.

Unnamed Stream near Walong (16). Unnamed small creek
about half a mile southeast of Walong (Caliente quadrangle), on
the Southern Pacific Railroad, was reported by Elmer Lyne of
Tehachapi to have been flowing about 10 gallons per minute prior
to the quake and some 100 gallons per minute about the middle of
August.

Clear Creek near Bealville (17). It appears that the headwa-
ters of Tehachapi Creek did not show an increase in the same
proportions as the downstream area. By contrast. Clear Creek, a
lower tributary, dry before the earthquake, started to flow within
abo\it 3(3 hours and was estimated at .3 to 4 cubic feet per second
by W. M. Jaekle of the engineering department of the Southern
Pacific Company. On January 19, 19.53, a current meter measure-
ment by T. A. Cooijer of the U. S. Geological Survey showed 2.6
cubic feet per second in Clear Creek, just upstream from the
Southern Pacific Railroad high fill near Bealville (Caliente quad-
rangle). On July 21, 19.53, the flow of Clear Creek at the same
point was estimated as 0.5 cubic foot per second.

Tehachapi Creek near Caliente (IS). L. E. Williams of Cali-
ente reported that water of Tehachapi Creek, reinforced by Clear
Creek. I'eached the confluence with Caliente Creek at Caliente
(Caliente quadrangle), about ,July 2(') and that both Caliente Creek
and Tehachapi Creek reached their maximum summer flow at
Caliente ab(mt August 10. Since the record of the V. S. Bureau
of Reclamation shows the flow holding steady at 21 second-feet
below the confluence, August 20-29. it seems rea.sonable to estimate
the flow of each stream, I'eported as similar in size, above the con-
fluence, at about 12 second-feet on August 10. Tehachapi Creek
was down to 0.5 cubic foot per second on July 21, 19.5.3, as meas-
ured by T. A. Cooper.

Indian Creek at Lorraine (19). C. W. Poole, the U. S.
Weather Bureau observer at Lorraine (Mojave qimdrangle), at
the confluence of Indian and Caliente Creeks, reiiorted as follows:
Both creeks had been dry July 1-20. 19.52, preceding the earth-
quake. During the latter part of the preceding winter (1951-52)
Caliente Creek above Lorraine had flowed for the first time in
four years. Indian Creek had flowed as usual during the same
winter (1951-.52). Indian Creek started to flow at Lorraine 1 week
after the earthquake. Most of the water originated in a southeast
fork of the creek, on the southwest shoulder of Cache Peak. By
the latter part of August, Indian Creek was carrying an estimated
3 cubic feet per second. It is notable that Caliente Creek did not
start to flow upstream from Indian Creek at Lorraine until after
the rains of Xovember 1952. Mr. Poole offered as an explanation
that there are large areas of sand in the upper Caliente Creek
channel, and that percolation slowed the advance of surface flow.

Mr. Poole reported no rainfall at Lorraine from April 19.52 until
early in Xovember 19.52. However, there were thunderstorms in
some neighboring areas late in July and in August.

Stndhorse Creek (20). Jlr. Poole stated that Studhorse Creek
(Mojave quadrangle), first tributary to Caliente Creek down-
stream from Indian Creek, started to flow iiuniciliately after the
quake.

I'nnunied Stream near Caliente (21). An unnamed small creek
nearly o|)posite Devil Canyon (Caliente quadrangle), tributary to
Caliente Creek, starte<l to flow soon after the earthquake. It is
believed that a small flow also appeared in Devil Canyon.

Oiler Canyon Spring (22). A spring appeared .-ibingside the
road in Oiler (\inyon (Caliente quadrangle), immediately after
the quake, and was still running in January and February 1953.

Part I]

Geology

Rock f^pring (23). Rook SpriiiK. nboiit 1 mile northeast of
Caliente (Caliente quadrangle), on the L. K. Williams Kanch. was
reported to have increased in flow from 'A gallons jier minnte before
the (jnake to (i afterwards.

Caliente Cictk ahove ami below Tehaehnpi Creek at Caliente
(24 and 2."i). Tehachapi Creek joins Caliente Creek at the small
town of Caliente (Caliente qnadrangle), anil in the words of the
local people, "dry as popcorn" described both creeks for at least
several weeks precetling the earthquake. Springs in the headwaters
of the creeks started to flow immediately after the quake on July
21 and Caliente Creek water reached the point of confluence about
July 25, followed by Tehachapi Creek water about a day later.
The flow in both creeks continued to increase as water from newly
flowing tributaries saturated the channels and came through.
Caliente Creek below the confluence is believed to have reached
its niaxinuun about August 10 with a flow estimated at about 25
cubic feet per second, and then held fairly constant until the
Xovember rains increased it.

Slightly more than a month after the Arvin-Tehachapi earth-
quake the V. S. Bureau of Reclamation, on August 26, 1952,
established a gaging station on Caliente Creek below its con-
fluence with Tehachapi Creek. The tributary drainage area is
about .34(1 square miles. The record obtained at this station, to-
gether with the estimate made by local residents, has been plotted
on figure 7 to form a hydrograph for the period of July through

PO 15 20 25

July

10 15 20 25
August

10 15 20 25

September

Figure 7. Hydrographs of daily discharge for Caliente
Creek at Caliente.

September 19.52. All of the runoff occurring subsequent to July
21, 19.52. as shown on figure 7. is assumed to have originated
from the earthquake since there is insufficient data to develop a
clear-cut delineation between the flow originating because of the
earthquake and that due to reduced evapotranspiration losses and
precipitation. The estimated runoff for the period .Inly 26 to Sep-
tember 30. 1952 amounts to about 3,(K)0 acre-feet for Caliente
Creek below the confluence with Tehachapi Creek.

On July 21, 1953, exactly 1 year after the major earthquake,
T. A. Cooper measured 3.3 cubic feet per second in Caliente Creek
at Caliente. Of this amount only 0.5 cubic foot per second was
contributed by Tehachapi Creek. Since there is normally no flow
at Caliente at this time of year it appears that the flow in Caliente
Creek and especially in the main stem ahove the confluence with
Tehachapi Creek is still .showing notable results of the 19.52 earth-
quake. In fact, the increase in runoff, due to the earthquake, for
Caliente Creek above Tehachajji Creek would be in the order of
5,000 acre-feet for the year following the quake.

Caliente Creek near Bena (26). Caliente Creek water ad-
vanced downstream from Caliente at a relatively slow rate since
large quantities of water were required to wet the channel suf-
ficiently to sustain surface flow. About 3 miles upstream from
Bena (Caliente quadrangle), a railroad station alongside State
Highway 466, there is a wide channel area that is normally
swampy. Water was observed there in ponds during January and
February 19.53. This area was reported practically dry preceding
the quake. Water is believed to have reached it about August 1,
19.52 and to have saturated it sufficiently by the middle of the
month to permit surface flow to proceed downstream.

Water reached the bridge on State Highway 4(!0 near Bena on
August 24, as shown by the water-stage recorder record of the

r. S. Bureau of Reclamation. From that point downstream,
Caliente Creek channel widens out into a sand and gravel delta
where percolation is extremely rapid. About 3 miles downstream
from the bridge on State Highway 4(;(i, and downstream from
the confluence of Walker Hasin Creek and Caliente Creek, the
creek is crossed by Xeumarkel Uoad. The ford at this point is
usually dry on the surface, and impassible on only very rare
occasions.

A. F. Xeumarkel stated that surface flow reached the ford
about Xovember 10. 1!».52, after fall rains began. The ford was
impassable for automobiles for about a Aveek following the heavy
rains of Xovember 14 and 15. Flow continued intermittently ail
winter and spring. For long periods there would be surface flow
only during the night. Following the rains of May 27 and 28, 1953
the creek again was not fordable for two days. Flow then decreased
rapidly and surface flow at the ford ceased entirely on June 1, 19.53.

Surface flow at the highway crossing near Bena lasted until
about June 12. T. A. Cooper reported that both Caliente Creek
and AValker Basin Creek were dry at the highwav bridges near
Bena, on July 21, 19.53.

^kunk Spring. White Wolf Ranch (27). Fred W. Xighbert
stated that this spring, high up on the northwest side of Bear
Mountain (Caliente quadrangle), practically doubled its flow im-
mediately after the quake of July 21. He also stated that at a
lower elevation he had been pumping water, since the earthquake,
from a well originally drilled for oil.

Unnamed Streams at Bear ^[ou^tain (28 and 29). Two un-
named creeks on the northwest side of Bear Mountain (Caliente
quadrangle), crossing the White Wolf fault, were observed to be
flowing on February 17, 19.53 and were reported by L. B. Krauter
of Tehachapi to have started flowing early in August following
the quake. It is probable that some of the water in Xo. 28 origi-
nated at or near Skunk Spring, Xo. 27.

While Wolf Springs (30). White Wolf Springs (Caliente
quadrangle), about 2 miles west of White Wolf Ranch headquar-
ters, is stated to have shown a definite increase in flow after the
quake.

Rock Pile Spring (31). Rock Pile Spring, 4i miles west of
White Wolf Ranch headquarters (Caliente quadrangle), was re-
ported to have gone dry at the time of the July 21 quake but
started to flow again following the heavy shock of August 22.

Berenda Spring (32). The Berenda Spring is located about
4 miles northwest of White Wolf Ranch headquarters (Caliente
quadrangle). Arthur J. Xeumarkel and Fred W. Xighbert agree
that it flowed strongly until July 21. 1952 and was used for stock
watering. There has been no flow since that date.

Xeumarkel Spring (33). The Xeumarkel Spring is about 7
miles northeast of Arvin (Caliente quadrangle). Arthur F. X'eu-
markel stated that it "flowed strong" until the earthquake of 1906
when it decreased so much that it required pumping. It was
pumped until 19.38 when the water was still "reachable" but pump-
ing was discontinued. After the earthquake of July 21. 1952 the
water "went out of sight."

Walker Basin Creek at Walker Basin (34). Leroy Rankin of
the Rankin Ranch in Walker Basin (Caliente quadrangle), stated
that the creek at the lower end of the basin is perennial but that
within a few days after July 21 the flow increased from a trickle
to some 2 or 3 cubic feet per second. This was a direct result
of new and increased flow in tributary springs, especially along
the w-est and north sides of the basin. Downstream from Walker
Basin, in the canyon section, new springs also appeared and many
old ones increased their flow.

Tributaries to Walker Basin Creek (35-41). These springs and
small creeks along Walker Basin Creek (Caliente quadrangle),
were all reported by Leroy Rankin to have started or increased
their flow very shortly after July 21. Nos. 35, 37, and 38 were
reported to be new springs. The largest was X'o. 40, Benninger
(Ca.stro) Canyon, for which the flow about the first of August
1952 was estimated at 700 gallons per minute.

Springs near Walker Basin (42-49). Increase in the flow of
these springs (except Xo. 43) in the mountains between Walker
Basin Creek and Kern River (Caliente quadrangle), was ahso
reported by Leroy Rankin. Xo. 43 ceased to flow. X'o. 46 (Fig
Tree Canyon) is distinctive because^ it dried up for a period of
about 2 weeks following the quake and then opened up again with
a flow greater than before. It was back to about normal in Febru-
ary 1953.

Earthquakes in Keen County, 1952

[Bull. 171

'•tttJ^

.J-^

m/f//'

V ■" ■'■'■■ -^2pv

W ■: \ , ;' '"■.

-s

.''

%■'■ ••'■.'

;V / '■■■-' •■/■ ;' '

tt/

-!_1

i'.,,, ■ ; _ ' ?

■/\

'////////,

1 '

c-j •

■"'»; -.;"'

1 i-Vii

• :

^■■■■■/■. ' I "l

;

I ' '

;i^^

I ;

■1 . !^

'■

rl
'',"

'■'.

H^i

5 10 15 20 25 30 5 iO IS 20 25 31 5 10 l5 20 25 31 5 lO 15 20 25 30
June. July August September

^.increose

in cJischorge due to
Tetiochopi eofthquohe

1 till

Figure 8. Hydrographs of daily discharfie for Walker Basin
Creek at Indian Mill Hock.

Walker linsin Creek at Indian Mil! Rock (50). The V. S.
Bureau of Reclamation ha.<i maintained, for several years, a gaging
station on Walker Basin Creek near the lower end of the canyon
at Indian Mill Rock about 2 miles northeast of Bena (Caliente
quadrangle). Tributary drainage area is about 111 square miles.
The record of daily discharge obtained at this site for the period of
June through September 1!).~>2 is shown on figure 8. This record
indicates that the stream channel became dry on July 8 and
remained dry until about four days after the earthquake. Ante-
cedent records show the streambed to be dry throughout most of
the summer months. The records obtained in 1047, and also shown
on figure 8, indicate that the daily discharge can change rapidly
in short intervals of time for many reasons other than earth-
quake. Because of this extreme variability in discharge, it is diffi-
cult to attempt a segregation of the Arvin-Tehachapi earthquake's
effect on the flow.

The lower part of figure 8 gives the best estimate available at
this time of the increase in discharge due to the Arvin-Tehachapi
earthquake. The maximum increase in discharge amounted to 4.1
cubic feet per second on July 30 and decreased to about 3 cubic
feet per second on September .30. This total increase in flow attrib
utable to the earthquake amounted to about 440 acre-feet, for the
period July 25 to September 30, 19.52.

Walker Basin Creek near Bena (51). The channel of Walker
Basin Creek passes under State Highway 4(56 about 1 mile west
of Bena (Caliente quadrangle) and perhajis 4 miles, by stream
channel, downstream from Indian Mill Rock. This 4-mile reach
lies in an absorptive gravel-and-sand formation and the advance
of the surface water was very slow. Although no definite informa-
tion is available it is believed that it was near the end of August
before the flow reached the bridge on State Highway 466. Water
was still flowing there in January and February 19.5,3. It ceased
flowing sometime before July 21, 1953.

I'ascoe Kprinu (.52). The Cecil I'ascoe household spring was
located in Caldwell Canyon about 2h miles northeast of the former
Kcrnville (Kernville quadrangle), and about 1 mile southeast of
New Kernville. It ceased flowing on July 21, 1952 and was still
dry one year later. However, new small springs broke out in the
vicinity, and the M. L. Crowder spring, half a mile west, increased
its flow about 50 percent.

South Fork Kern River near Onyx (53). The discharge of the
South Fork Kern River at the southern extreme of the Sierra
Nevada has been measured for many years by the I'. S. Geolog-
ical Survey at a site 5 miles northeast of Onyx (Kernville quad-
rangle). The contributing drainage area is 531 square miles. The
records have been published in the annual water-supply papers,
and the hydrographs on figure 9 give daily discharge during 1938,
1941 and 1952 for the four-month period of June through Sep-
tember. Records for the 2 earlier years were selected for inclusiou
on figure 9 because of similarity in discharge at the beginning of
the summer recession period. The slopes of the hydrographs for
all 3 years are quite similar through June and July until modified
by the summer rainfall.

In 1952 the normal recession was interrupted by a pronounced
increase in discharge on .Itily 26. This increase in flow is believed
to be more closely associated with the summer rainfall occurring
at that time, than the Arvin-Tehachapi earthquake. However, the
well-sustained flow of 149 to 155 cubic feet per second between
July 27 and August 1 suggests that the earthquake may have
had some influence on the runoff.

Hot Springs near Bodfish (54). The Scovern Hot Springs, 2
miles northeast of Bodfish (Kernville quadrangle), spouted a con-
siderable increase of flow at the time of the earthquake, July 21,
19.52. A similar action occurred with the quake of March 15, 19.33.
On July 21, 1953, 1 year after the recent quake, the flow was
estimated at 0.25 to 0.30 cubic foot per second, still somewhat
more than normal flow.

Bodfish Creek at and near Bodfish (53). Frank Laurence of
Bodfish reported that the creek at the town of Bodfish (Kernville
quadrangle) was dry on July 21, 19.52. At a point about 3 miles
upstream it started to flow on .July 22 or 23 but the flow did not
reach town during the summer of 1952.

On March 17, 19.53 T. A. Cooper of the U. S. Geological Survey
measured 0.6 cubic foot per second in Bodfish Creek, three-quarters
of a mile downstream from the town. It is concluded that the earth-
quake had little lasting effect upon the flow of this creek.

A small spring on the Laurence property in Bodfish almost
doubled its flow within a few hours after the quake of July 21.

Democrat Springs (56). The hot spring which supplied the
plunge at Democrat Springs in Kern River Canyon, 10.5 miles
southwest of Bodfish (Tobias Peak quadrangle), ceased flowing
entirely on July 21, 1952. During April 1953 it showed a slight
trickle and by July 21, 1953 it was flowing 8.0 gallons per minute.

A'ern River near Democrat Springs (57). A gaging station with
a recording instrument has been maintained a mile downstream
from Democrat Springs (Tobias Peak quadrangle) by the Southern
California Edison Co. (under direction of the I'. S. Geological
Survey), since 1951. Tributary drainage area measures 2,264
square miles.

The record of daily discharge showed no unusual change on
July 21-22 but did show an increase in discharge on July 26 and
27, 1952 similar to that shown by South Fork of Kern River near
Onyx (No. 53 above). For the same reasons cited there, it is con-
cluded that the increase was caused by precipitation and not by
the earthquake.

Kern Canyon Creek (.58). A small unnamed creek flows into
the Kern River on the left (south) bank, just downstream from
the mouth of Kern Canyon (Caliente quadrangle). Employees at
the nearby power-plant reported that the creek started to flow soon
after July 21, 1952. It was dry at the same point on March 16,
1953.

Cottonwood Creek near Kern Canyon (59). Leroy Rankin, who
grazed cattle in the upper Cottonwood Creek area (Caliente quad-
rangle) reported that many tributary springs increased their flow
shortly after the earthquake.

Joe Prowell, at the Kern Rock Co. plant on Cottonwood Creek,
about 1 mile upstream from its confluence with Kern River, stated
that the creek had stopped flowing prior to July 21, 1952. It
normally ceases flowing at that point about .Tune of each year and
does not start again until after the onset of winter rains. How-
ever, the creek started to flow at the rock company plant about
July 25, 1952, and continued throughout the balance of the sum-
mer ami winter and on March 17, 19.53 was carrying 1.4 cui)ic
feet per second, as measured by T. A. Cooper of the U. S. Geo-
logical Survey. On July 21, 1953 he reported no flow.

White River near Ducor (60). A regular gaging station with
a recording instrument has been maintained on this stream eight
miles southeast of Ducor (White River quadrangle) since 1937.

Tart II

Geology

1500
lOOO

— I — 1 — 1 — I — 1 i ' ' — I 1 — 1 —

1 , 1

800

^% \-

-■J \

<^,.

M^-

\l^

UN. J

/ (

|80

' 40

rja:vr>]

_/■'■,

1
1

•-, i

A^"l

>
30

vi;

;

1 1 1 1 1 MM

i 1 i

1000
800

600
500
400

300

ZOO

ISO

100

60
50

20
15

10 15 20 25 30 5 to 15 20 25 30 5 10 15 20 25 31 5 10 15 20 25 30

June July August September

Figure 9. Hydrographs of daily discharge for South Fork
Kern River near Onyx.

The drairage area is 120 square miles. Records of daily discharge
are published in the U. S. Geological Survey annual water-supply
papers. This record shows a sudden increase on July 29 and 30.
followed by a rapid recession to no flow on August 5. Since the
increase corresponds to the thunderstorm period in that area, it is
concluded that it was due to the precipitation and not to the
earlier earthquake.

Santa Barbara, Ventura and Los Angeles County Areas

The following text de.seribes points of observation and
the effect of the Arvin-Tehaehapi earthquake on the
flow in streams and springs of Santa Barbara, Ventura
and Los Angeles Counties at stations indicated on figure
2. As already indicated, these data were obtained as a
part of the overall program of the U. S. Geological Sur-
vey in cooperation with the State of California, the U. S.
Bureau of Reclamation. Santa Barbara County Water
Authority, Ventura County Water Survey and Los An-
geles County Flood Control District.

Huasna River near Santa Maria (61). Huasna River, a tribu-
tary to Cuyama and Santa Maria Rivers, drains the south and
west sl(i|>es of the Santa Lucia Range. The runoff is measured at
an altitude of 600 feet at a site almut 0.5 mile upstream from
the stream's confluence with Cuyama River and 8 miles northeast
of Santa Maria. The runoff from this 119-s<|«iarp-mile drainage
area has been obtained since December 1929.

The runoff records at this site do not show any change in dis-
charge, subsequent to July 21. 1952, that could be attributed to
the Arvin-Tehachapi earthciuake.

Alamo Creek near Santa Maria (62). Alamo Creek, a tribu-
tary of Cuyama River, drains the low mountain areas between
the Santa Lucia Range and the San Rafael Mountains. The run-
off from this 87.7-.square-mile drainage area has been measured
since October 194.3 at a site about .580 feet above sea level and
1.2 miles above the confluence with Cuyama River, 9 miles north-
east of Santa Maria.

The runoff observations indicate that the discharge of this
stream was not affected by the Arvin-Tehachapi earthquake.

Tepiisquet Creek near Sisguoc (63). Tepusquet Creek, a trib-
utary to Sisquoc River and through it to San^a Maria River,
drains the south face of the San Rafael Mountains. A gaging sta-
tion was established in October 1943 to measure runoff at a site

about 500 feet above sea level, 3 miles east of Sisquoc. The stream
has a drainage area of 28.9 square miles at this site.

The discharge records indicate that the Arvin-Tehachapi earth-
quake did not have any measuralile influence on the runoff of this
stream.

Sisquoc River near Sisquoc (64). Si.squoc River, a tributary
of Santa Maria River is an east-west stream draining the interior
areas of the San Rafael Mountains. The runoff from this stream
system has been measured between December 1929 and September
19.33. and since October 1943 at a site where the stream discharges
onto the alluvial valley floor aliout 7 miles east of Sisquoc. The
station is located at about 620 feet above sea level and measures
the runoff from a 290-square-mile area.

Runoff measured at this station since July 21. 1952 does not
indicate any noticeable change in discharge attributable to the
Arvin-Tehachapi earthquake.

Sahipuedca Creek near Lompoc (65). Salsipuedes Creek, a
tributary to Santa Ynez River, drains the interior regions of the
westerly end of Santa Ynez Mountains. Runoff from this 46.6-
square-mile drainage area has been measured since January 1941
at a site on the Jalama Road bridge, just downstream from El
Jaro Creek and about 5 miles southeast of Lompoc. The altitude
of the gage is about ,340 feet above sea level.

Runoff records obtained at this site since July 21, 19.52 do not
indicate any measurable change in discharge attributable to the
Arvin-Tehachapi earthquake.

Juan Y Lolita liancho Spring (66). This .spring is on the
north side of the Santa Ynez Mountains at an altitude of about
1,1.50 feet, about 1 mile north of the Santa Ynez fault. It is on
the Juan Y Lolita Rancho about 3 miles south of the town of
Santa Ynez. Monthly observations from February 1. 1949 to Janu-
ary 9. 1952 indicate a range in discharge of 0.9 to 3.3 gallons per
minute. As a result of a fairly wet winter, the flow from this
spring increased to 50 gallons per minute on April 16, 19.52. then
decreased to 9 gallons per minute on July 3. the last observation
prior to the Arvin-Tehachapi earthquake.

Immediately after the earthquake, the flow increa.sed, reaching an
oKserved maximum of 17 gallons per minute on October 21. 19.52.
Flow was then sustained at 13 to IS gallons per minute throughout
the dry 19.52-.53 winter, declining to a flow of 12 gallons per
minute on June 1, 19.53. This gain in flow attributable to the
Arvin-Tehachapi earthquake amounted to about 16 acre-feet by
June 1, 19.53.

Juan Y LolHa Rancho Spring (67). A second and somewhat
smaller spring on the Juan Y Lolita Rancho is located almost in

Earthquakes in Kern County, 1952

[Bull. 171

—

Mncrea
Ydischo

se inT/V
rge due

in '/////y

1953

Figure 10. H.vdrosraplis of (U.seliarKe for .Tuaii V I.olita
Raiu'ho ."iprinK.

the Santa Ynez fault zone and about 1 mile south of station (66).
This spring is also on the north side of the Santa Ynez Mountains
at an altitude of aliout l.lfjO feet. The very noticeable ehanse in
flow of this spring caused by the Arvin-Tehachapi eartluiualie is
shown graphically on figure 10. This diagram gives the monthly
observations of How made at this station from February l!t49 to
June 1953. The hydrograph, developed by connecting Successive
points of measured discharge by straight lines, indicates a succes-
sively decreasing annual cycle till the winter of lOf)! due to de-
pleted mountain ground-water storage. Then, as a result of a
substantial recharge during the fairly wet winter of l!).")l-5t2. the
discharge increased to I?..") gallons per miuute on April 14, 1952.
Sub.sequently the flow diminished to 1.25 gallons jier minute on
July 3, 1952 just prior to the earth(puiUe.

The first measurement after the Arvin-Tehachapi earthqual;e
indicated an increase to 9.0 gallons per minute on September 2,
1952, then a decrease to 3.0 gallons per minute in June 19.53.
Thus at the end of this period whicli followed the very dry 1952-
53 winter, the ol)served flow was almost as great as that shown
for April in the previous wet year.

It has been estimated tliat had tlie earlhciualie not occurred, the
flow would have followed the iiatteru shown by the dashed line.

Figure 11.

Hydrographs of discharge for Caiiada del
Refugio Creek.

The difference between these two hydrographs, shown by the
cross-hatched area, represents the estimated increase in flow at-
tril)utable to the Arvin-Tehachapi earthquake. This gain in run-
off is shown to a much better advantage in the lower part of
figure 10.

J. V. Crawjord Spring (68). This spring, also located on the
north side of the Santa Ynez Mountains at an altitude of about
1,7.50 feet, and close to the Santa Ynez fault, showed an appre-
ciable increase in flow as a result of the Arvin-Tehachapi earth-
quake. The monthly ol)servatious showed about the same pattern
of runoff distribution as that shown on figure 10. During the
winter periods of 1949, 19.50, and 1951 the flow ranged from
about 0.4 to 0.8 gallons per minte. Then, due to the greater re
charge during the 1951-52 winter, the flow increased to 1.5 gal-
lons per minute. Ry July 14, 1952 it had declined to 1.03 gallons
per minute.

With the advent of the Arvin-Tehachapi earthciuake, flow in-
creased to 1..50 gallons per minute on Septemlier 3, 19.52 and re-
mained in excess of 1.2 gallons per minute through .June 19.53.

irons Creek at Walska Estate (69). Wons Creek also origi
nates on the north side of tlie Santa Ynez Mountains and is tribu-
tary to the Santa Ynez River. The runoff from this very small
drainage area of aliout half a square mile is measured at an alti-
tude of 2,200 feet, about half a mile north of the divide. This
highland area is south of the Santa Ynez fault.

ilouthly observations at this site represent the composite run-
off of many individual .springs. The records show the same annual
cycle as sliown f)n figure 10 with greatest discharge in winter
and spring, and minimum discharge in summer and fall. A suli-
stantial portion of this runoff depletion during summer and fall
is due to evapotranspiration losses within the ari'a. Due to lack
of any substantial recharge to mountain ground-water storage,
winter runoff showed the same progressive decrease from 1949 to
1951 as indicated on figure 10. Then as a result of the substan-
tial ground-water recharge during the fairly wet winter of 1951.52,
winter runoff increased to 2t)() gallons per minute from the usual
winter runoff of 30 to 60 gallons per minute. Ry mid-June flow
had decreased to 104 gallons per minute.

Following the Arviu-Teliacbapi earthquake, flow increased to
132 gallons per minute on August 5, 1952 and remained in excess
of 47 gallons per minute through June 19.53. As a result, it has
been estimated that the excessive runoff attributable to the earth-
quake will amount to 67 acre-feet or about 2.5 inches of water
over the drainage area.

West Fork Qiiiota Creek at Forest Service Spring (70). West
Fork Quiota Creek is anotaer small highland drainage just north
of the diviile of the Santa Ynez Mountains and .south of the Santa
Yiu'z fault. Monthly observations of flow are made at an altitude
of 2,(»40 feet.

Between the spring of 1949 and the fall of 1951 this stream
channel was dry most of the time. However, as a result of the

Part 11

Geology

!)3

Krounil w.iti'i- ri'ilinr;:!' in tlio ltl51.r)2 rainy scasmi. flow nil .Iiiiu'
l,s, l!tr>;i was still 0.42 galli)ns por niiiintc.

Immi'diately folliiwins; llii' i'arth(|iial<('. tliiw iiicri-ascil to !."> gal
Ions ppr niinnte on August 5. l!ir>2. and the creek was not re-
ported dry until October 2(t, \U't'2. This latter small increase in
runoff is believed to be attributable to the Arvin-Tehachapi
earthquake.

('niiiidd del Refugio Creek near I'rfiK/io diinid Stnlioii (71)
Cafiada del Refugio Creek drains a steep frontal mountain area
on the south side of the Santa Ynez .Mountains and is tributary to
the Pacific Ocean. Monthly observatiou.s of flow are made at an
altitude of 4tK) feet just below the confluence of the two principal
forks.

The observatiou.s prior to the 19.")1T>2 rainy season were plotted
on figure 11 and indicate the same typical trend shown on figure
10. As a result of the fairly wet winter of li).")l-."i2. How increased
to r>(50 gallons per minute, then diminished to l.jO gallons per
minute on June 18, 1!C)2.

With the advent of the Arvin-Tehachapi earthquake, flow in-
creased to ^ZO gallons per minute on August 5, 19.">2 as shown
on figure 11. From that date until .Tune 19.5.''i the flow was sub-
stantially higher than normally to be expected during this year.
That portion of the runoff attributable to the earthquake is indi-
cated as in the preceding diagrams. This increase in the 11-month

Figure 12

1952 1953

Hydrographs of discharge for Carneros Creek.

period of August 1952 through June 1953 amounted to 139 acre-
feet. On an areal basis, this is equivalent to 1.2 inches of water
over the drainage area.

Carneros Creek (Bartlett Canyon) near Goleta (72). Carneros
Creek, like Canada del Refugio Creek, is a frontal stream on the
south side of the Santa Ynez Mountains. Monthly measurements
are made in Bartlett Canyon about 500 feet above sea level, 3
miles northeast of Goleta, and 5 miles upstream from the ocean.

All observations since July 1948 are plotted on figure 12. The
hydrograph developed from these observations shows the typical
general decline in flow due to a depleted ground-water storage
prior to the winter of 1951-52. The greater recharge resulting
from the 1951-.j2 precipitation cau.sed an increa.se in runoflf which
continued until the time of the Arvin-Tehachapi earthquake.

As a sequence to the earthquake, the flow greatly increa.sed and
was sustained well above the 1951-52 winter runoff. The estimated
increase in the flow attributable to the earthquake has been
cross-hatched in the hydrograph and replotted in the lower part
of the diagram. During the 11-month period subsequent to the
earthquake this additional runoff is believed to be in the order
of 60 acre-feet and is equivalent to 0.6 inch of water over the
drainage area.

Canatsey-O'Bannon Spring (73). This spring consists of a
seep from landslide material on the canyon wall. It is at an alti-

■5 80

1 Eorlhquoki 1

-^^

^ ' a.

/^-^.^

^v.

-OO'X'O

''<'\^cx/^

V>V"

Figure 13. Hydrographs of discharge for Mrs. K. C. Cauatsey
and Mrs. B. O'Bannon spring.

tude of 425 feet in the basin of San Pedro (or San Piedro) Creek,

a frontal stream on the south side of the Santa Ynez Mountains,
and tributary to the Pacific Ocean. Monthly measurements are
made about 3 miles north of Goleta and 4 miles upstream from
the ocean.

The monthly observations shown on figure 13 indicate a re-
markably constant flow prior to the Arvin-Tehachapi earthquake.
This uniformity of How was abruptly disrupted by the earth-
quake. In a very short interval of time flow increased from about
30 gallons per minute to 125 gallons per minute on July 30. Sub-
sequent to this date, the flow declined steadily to S3 gallons per
minute in ,June 1953.

The cross-hatclied area on this hydrograph and the one plotted
in the lower part of figure 13 indicate the estimated increase in
runoff attributable to the earthquake. In the 11-month period of
August 19.52 through June 19.53 this increased runoff has
amounted to 116 acre-feet. This spectacular increase is more than
2.5 times the entire annual runoff prior to the earthquake.

Holmes Spring (74). This is the first of a series of springs in
the frontal drainage area of San Jose Creek, on the south side
of the Santa Y'nez Mountains. This spring is in the form of a
seep from the alluvial stream bed deposits at an altitude of about
1,900 feet.

Observations made prior to the occurrence of the Arvin-Tehach-
api earthquake have the .same general uniformity of How shown
on figure 13. A seepage of 2.4 gallons per minute on June 3, 1952
increased to 14 gallons per minute on August 4. Subsequently the
flow diminished but remained above 6 gallons per minute prior to
June 1953.

San Jose Creek at Holmes' Place near San Marcos Pass (75).
Monthly observations are made in this small headwater drainage
area at an altitude of 1,890 feet, about 5.5 miles north of the
ocean. A hydrograph developed from these observations is given
on figure 14. In general, the hydrograph prior to the Arvin-
Tehachapi earthquake is similar to those shown in the preceding
diagrams. Subsequent to the earthquake the discharge increased
greatly and was sustained through June 19.53. That portion of
the runoff believed to be attributable to the earthquake is shown
by cross-hatching and replotte<l for better definition in the lower
part of figure 11. During the 11-month period subsequent to the
earthquake and ending in June 19.5.3, this e.xcessive runoff has
been estimated as 55 acre-feet or the equivalent of 0.86 inch of
water over the drainage area.

San Marcos Trout Club Spring near San Marcos Pass (76).
The San Marcos Trout Club Spring has its immediate origin in
the alluvial deposits in San Jose Creek at an altitude of about
1,700 feet, about 1.5 miles southeast of San Marcos Pass. Monthly
observations at this site since July 1948 indicate a flow that

Earthquakes in Kern County, 1952

[Bull. 171

Figure 14. Hydrographs of iliscbarKe for San Jose Creek at
Holme's place near San Marcos Pass.

generally ranged from 2 to 6 gallons per minute, except for short
winter periods, prior to the Arvin-Tehachapi earthquake. Subse-
quent to the earthquake the flow increased to !> gallons per minute
and remained in excess of 6 gallons per minute through .Tune
1953, along a pattern very similar to that shown on figure 14.

Unnamed Tributary to San Jose Creek at Hobo Rock near San
Marcos Pass (77). This unnamed tributary to San Jose Creek
is measured monthly at a site having an altitude of about 1,70()
feet, about li miles east of San Marcos Pass. During the earlier
part of the record the distribution was quite similar to that shown
on figure 11. The discharge reflected the depletion of the ground-
water. (Jround-water recharge in the l!tr)l-r)2 rainy season in-
creased the runoff and gave a sustained flow prior to the earth-
quake comparable to the earlier winter runoff. With the advent
of the earthquake, the flow increased but was not as well sus-
tained, as shown on figure 11.

San Jose Creek (78). San Jose Creek is measured monthly
at a site 1 mile above Patterson Ave. Bridge, at an altitude of
about 2.~)0 feet. Discbarge followed a pattern very similar to that
shown on figure 14 prior to the occ\irrence of the Ar\in-Tehachapi
earlbqual;o. Sulisecpient to the earth(iuake, flow increased from
aliout S(t galhjns per minute to 280 gallons per minute. This
increase in runoff attributable to the earth(piake was not as well
sustained as for some of the measuring jioiuts upsti'eam.

Cold Spring Canyon Creek near San Marcos Pass (79). Cobl
Spring Canyon Creek is on the north side of the Santa Yncz
Mountains and is tributary to the Santa Ynez Hiver. Monthly
observations of the flow in this headwater drainage area are
made about 1,(500 feet above Kea level.

Like many of the i)receding records, the flow indicateil ;i
gradual decreasing trend prior to the winter of 19ril-r)2. During

that winter period there was an appreciable increase in discharge,
followed by a recession that produced a runoff of 28 gallons per
minute on July 15, 1952. With the advent of the Arvin-Tehachapi
earthquake, flow increased to G4 gallons per minute on September
4, 1952 and was well sustained through .Tune 1953.

Ifot Spritiys Creek (SO). Also located on the north side of the
Santa Ynez Mountains is Hot Springs Creek. Monthly measure-
ments are made about 9.50 feet above .sea level and about 1 mile
above the confluence with the Santa Ynez River. After a fairly
wet winter the discharge had declined to 17(i gallons per minute
on .July 15, 1952. Sul)se(|uent to the Arvin-Tehachapi earthquake
it increased to 300 gallons per minute on September 4, 1952,
declining to 128 gallons per minute on June 5, 1953.

Crown Eleven Ranch Spring (81). Crown Eleven Ranch
Spring is al.so located on the north side of the Santa Ynez
Mountains between Hilton and Tequepis Canyons, at an altitude
of about 990 feet. Monthly observations between October 1948
and December 1951 showed a range in flow of 0.2 to 1.8 gallons
per minute. As a result of the 1951-52 rainy sea.son, flow increased
to almost 3 gallons per minute and remained fairly well sustained,
declining to 2.3 gallons per minute on July 14, 1952. Subsequent
to the Arvin-Tehachapi earthquake, flow increased to 5.0 gallons
per minute, the highest discharge recorded during the period of
record, and remained sustained at this amount or more until
February 1953. The latest observation on June 2, 1953 indicated
a discharge of 3.6 gallons per minute.

Cachuma Creek near Santa Ynez (82). Cachuma Creek drains
the southern slojjes of the San Rafael Mountains and is ti Unitary
to the Santa Y'nez River. A gaging station was established in
October 1950 about 3.6 miles upstream from Santa Ynez River
and 8.8 miles east of the town of Santa Ynez.

:i^i'-"^-^

^-^-K

"sj

|-'-K

^^ S>

fhrf?

■K^/fTh.

^■l\

M/M/M

V////////////////////M

\*Tr\

\»

mm§,

"l\

m \'//////// ; \ ^-^

—

mf"

^^r^--.'

944^

\" 1

' 1 1 ^l 1 '"i ^ ^

_ 1 . L J _

5 10 15 20 25 30 5 10 IS 20 25 31 5 10 15 20 25 51 5 10 15 20 25 3

Increase ir. aischorie /f/TTTtn,,

r7 f//£

■

,,,,////////////////////

'////////}

Try.

mf^mm^MC ^~~~^

June

July August | September

Figure 15. Hydrographs of daily discharge for Santa Cruz
Creek near Santa Ynez.

The discharge at this station does not reflect any influence
from the Arvin-Tehachapi earthquake.

Santa Cruz Creek (83). Santa Cruz Creek drainage area is
also on the south and western slope of the San Rafael Mountains
and is adjacent to the Cachuma Creek drainage area. The gaging
station is about 0.5 mile above the stream's confluence with
Santa Ynez River. Unlike Cachunui Creek, runoff from this 77-
scpiare-mile drainage area increased immediately following the
Arvin-Tehachapi earthquake.

This station has a record of daily discharge beginning in
October 1941. The daily record for the 4-month period of June
through September is shown on figure 15 for 1952 as well as for
the two antecedent years of 1942 and 1944. Discharge increased
promptly after the earthquake, and reached a maximum in the
first part of August.

On the basis of antecedent records, it has been possible to
estinuite the How had there been no earthquake. The difference
between the dashed line and the 1952 record shows the increase

Part 1!

Geology

nttriliiitnlilc to the Arvin-Tehacliiipi oaithqviake. It reachi'd a
inaNiiiiiiiM of abovit S.!) oiil)ic feet i)er second on Aujjost 'A, 1!)52
an<l then (,'ra<luall.v decreased to about 0.6 cubic foot per second
on Seiitenilier 'M), l!ir>2. This increase in runoff subse(]uent to the
earthqualie aniounte<l to 2!)7 acre-feet. On an areal basis, this is
eiiuivalent to 0.07 inch over the <lrainaf;e area.

Cuiiamn Uiier near Vetituropn ( S-1 ) . Some distance ea.st and
slightl.v north of Santa Cruz Creel< drainage area, on the nortli
side of Pine Mountain, is the headwater area of Cuyama River,
a tributary to Sania Maria River. In Xovend)er 1044 a Kaning
station was established at an altitude of about ;?,r>t>0 feet at
t)zena, aliout 12 miles southeast of Ventucopa.

The record of daily discharge shows the .s.ame pattern of
distribution as that on figure 15. Discharge of 0.5 cubic foot per
second on .July 20, 1052 increased to .H.O cubic feet per second
three days later. The total increase in runoff attributable to the
Arvin-Tehachapi earthquake amounted to about ;i(X) acre-feet prior
to October 1. 10.52. This increase in runoff is equivalent to about
51 percent of the entire annual runoff during the dry 1051 water
year. On an areal basis it is equivalent to about t).()7 inch of
water over the drainage area.

■^

J95
194

■SnO

F>>^

"^V

^-N

h 1

a,.

. -^

:,.y^

S^^

aT^

-^vL^

1952

a
6

'^w^

m/M/m:..

■^

'-■--. — ^

•^^~.

^^^

\.^

■^

/±

-y

U^iUi

"■

J946

0 15 20 25 31 5 10 15 20 25 3C?
August September

1 1

argc
uoke

^§mi

to Arvi

) eorthq

w//////////////

1 ;

mSMMMM

i
1

■

f//////////,

W/////i

Figure 16.

Hydrographs of daily discharge for Matilija
Creek above Matilija Dam.

Matilija Creek above Reservoir (85). South of the Cuyama
River headwater areas and across the upper Sespe Creek drainage
area is Matilija Creek, a tributary of Ventura River. Matilija
Creek drains the mountain plateau north of the Santa Ynez Moun-
tains. The daily discharge is measured about 1,160 feet above sea
level and ab<uit 2 miles upstream from Matilija Dam. Records are
available since May 1048, and at a site 2.4 miles downstream since
October 1027.

The effect of the Arvin-Tehachapi earthquake was to increase
the runoff from many of the springs and streams in the area. As
shown on figure 16, daily discharge progressively increased from 10
cubic feet per second on July 20, 1052 to 21 cubic feet per second
on August 11 and 12, largely as a result of the earthquake.

The hydrographs on figure 16 include records obtained in 1041,
1943, and 1046 at the station downstream so that daily discharge
could be estimated subsequent to .July 20, 1052 had there been no
earthquake. Flow under these conditions is shown by a dashed line
and the cross-hatched area between these estimated and observed
discharges represents the increase attributable to the earthquake.

This increase in runoff prior to October 1, 1052 amounted to
1,170 acre-feet or only slightly less than the total annual runoff
during the preceding dry year of 1051. On an areal basis, it is
equivalent to 0.43 inch of water over the drainage area.

North Fork Mnlilijn Creek at Matilija (86). East of the
Matilija Creek drainage area (85) is North Fork Matilija Creek.
Records of daily di.scharge are available since 1028 at a site about
0.5 mile above its confluence with JIatilija Creek. There was a
pronounced increase in discharge immediately following the earth-
quake, along a pattern very similar to that shown on figures 15
and 16.

The increase attributable to the earthquake was about 560 acre-
feet, or the equivalent of about 0.68 inch over the drainage area.

Coyote Creek near Ventura (87). Coyote Creek, a tributary to
Ventura River, drains the south slopes of the Santa Ynez Moun-
tains and the foothill areas on the ocean side of these mountains.
A continuous record of discharge has been obtained for this stream
since October 1027, except for the period of October 1032 to Sep-
tember 1933, at a point about 0.2 mile above its confluence with
Ventura River and about .5.5 miles northwest of Ventura.

The records indicate that the <li.scharge declined continuously
throughout the 4-month period of .Tune through September 1952
without modification as a result of the Arvin-Tehachapi earthquake.

Santa Paula Creek near Santa Paula (88). Santa Paula Creek,
a tributary to Santa Clara River, originates on the south side of
Topatopa Mountains and on the north side of Santa Paula Peak.
A gaging station was established in October 1027 just upstream
from the Santa Paula Water Works diversion dam, about 3 miles
north of Santa Paula, to measure the runoff from this 39.8-square-
mile mountain drainage area.

The records obtained before and after the Arvin-Tehachapi earth-
quake do not indicate any definite change in runoff as a result of
this event.

Sespe Creek near ^y heeler Springs (80). The headwater areas
of Sespe Creek are on the southern slopes of Pine Mountain and
northern slopes of Ortega Hill and adjacent mountains. Discharge
is measured in Sespe (Jorge at a site about 3, .500 feet above sea
level and about 5 miles northeast of Wheeler Springs. Runoff from
this 50-square-mile drainage area has been measured continuously
since July 1048.

Daily discharge showed a general increase due to the ' Arvin-
Tehachapi earthquake. The runoff distribution was similar to that
on figure 16, increasing from 1.5 cubic feet per second on July 20,
1952 to 3.(i cubic feet per second on August 1.

As a result of the earthquake, the runoff was increased by 204
acre-feet, or the equivalent of 0.08 inch of water over the drainage
area.

Figure 17.

Hydrographs of daily discharge for Sespe
Creek near Fillmore.

Sespe Creek near Fillmore (90). This gaging station, down-
stream from the Wheeler Springs station, is about 0.1 mile down-
stream from Little Sespe Creek and 3.5 miles north of Fillmore.
At this site the stream channel has an altitude of only about .500
feet. Records of runoff from the 254-.square-mile drainage area have
been obtained continuously since 1934.

Earthquakes in Kern County, 1952

[Bull. 171

S ' 1 1

' \

]

, i

■"

-%Of ''

'„

1 1 i

S.V

.^^

iTiJ^i

HSI

\Nw7^<n^-

^v-^ /

U--^

■

"■"■■\

'\y>J94rV

'\ A

I9fi?

K*''/""'''

^-^

,rv^i

, ^--^'i

1942 " V^

: '"

"ip" X ^ ' '

•

1 1
1 1

5 tC 15 2C 25 iO 5 10 15 ZO Z5 3r 5 10 15 20 25 31 5 10 15 20 25 30"

MM 1 ' 1

r ~^^

Increose m drschofge due
\ to flf»in eo'lhquoke

1 1

""--.;„ '

m .......

June

July 1 August 1 Seplember

Figure 18. Hydrographs of daily discharge for Piru
Creek near Piru.

The discharge was definitely affected by the Arvin-Tehachapi
earthciiiake, a.s indicated on figure 17. Pi.scharge of 17 cubic feet
per second on July 20, 19ri2 increased to 37 cubic feet per second
on July lil largely as a result of the earthtiuake.

On the basis of records obtained in 1941, 1943, and 1944, it has
been possible to estimate the probable discharge at this station had
there been no earthquake. The cross-hatched area between the
estimated hydrograph and the observed discharge represents the
gain in discharges resulting from the earthquake.

This increase, plotted in the lower part of figure 17, ranged from
zero on July 21, 1952 to 22 culiic feet per second on August 1.
gradually decreasing to 9 cubic feet per second on September .31).
Thus, as a result of the earthquake, Sespe Creek acquired an addi-
tional runoff of 2,100 acre-feet. This is equivalent to 61 percent of
the entire annual runoff in the dry 19.")1 water year. On an areal
basis, the additional runoff is equivalent to O.IG inch of water over
the entire drainage area.

Hopper Creek near Piru (91). East of Sespe Creek is the
short frontal drainage area of Hopper Creek. The gaging station
is at the bridge on U. S. Highway 126, 2 miles southwest of Piru.
Records of daily discharge are available since 1930. Following a
normal summer recession, the stream channel became dry early in
July. Immediately following the Arvin-Tehachapi earthquake, the
stream started to flow, increasing to over 3 cubic feet per second
by mid-August along a pattern very similar to that shown on fig-
ure 16.

Piru Creek near Piru (92). East of Hopper and Sespe Creeks
is the 432-s(nuire-mile mountain drainage area of Piru Creek. Tliis
stream originates on the northern slopes of I'ine Mountain, and
after following a generally easterly course through the interior
mountain areas, turns sharply southward to join the Santa Clara
River. A continuous record of discharge is available for a site 1.8
miles northeast of Piru where the stream channel has an altitude
of about 7S0 feet.

As indicated on figure 18, <laily discharge was noticeably affected
by the Arvin-Tehachapi earthquake. The flow increased from 7..")
cubic feet per second prior to the earthquake to more than 16
cubic feet per second during the first part of August.

On the liasis of the antecedent records obtained in 1942, 1943,
and 1944, it was possible to estimate the probable discharge sub-
sequent to July 21, 1952, had no earthquake occurred. The cross-
hatched area between these estinuited records and the observed
data represents the gain in flow attributable to the earthquake.

This increase, as shown in the lower part of figure 18, amounts
to 920 acre-feet, or the equivalent of 0.04 inch of water over the
432-square-mile drainage area.

Santa Clara Hirer near Saugus (93). The headwater areas of
the Santa Clara River lie east of Piru Creek in the San Gabriel

Mountains. Continuous records of discharge are available from the
gaging station at V. S. Highway 99 crossing, about 3 miles west
of Saugus, since September 1929. The altitude of the stream chan-
nel at this site is about 1,040 feet.

The records indicate that the Arvin-Tehachapi earthquake did
not affect the flow at this point.

Santa Clara River near Lang (94). The gaging station is at
an altitude of 1,735 feet in the headwater area of Santa Clara
River about 0.7 mile east of Lang Railway Station. A continuous
record of discharge exists since October 1949.

The data obtained at this site do not reflect any change in flow
as a result of the Arvin-Tehachapi earthquake.

Little Rock Creek near Little Rock (95). Still further east, on
the north side of the San Gabriel Mountains, is the Little Rock
Creek drainage area. A continuous record of discharge is available
for this stream at an altitude of 3,290 feet, about 5 miles south of
Little Rock. This record is important because the station is located
about 5 miles upstream from the San Andreas fault zone.

The records at this site do not indicate any change in flow as
a result of the Arvin-Tehachapi earthquake.

Rock Creek near Valyermo (96). The Rock Creek drainage
area is just east of the Little Rock Creek drainage area and is
also on the north side of the San Gabriel Mountains. An almost
continuous record of discharge of this stream e.xists since January
1923 at the gaging station located about 1.8 miles .southeast of
Valyermo, at an altitude of about 4,0.50 feet. The important San
Andreas fault zone is within 1 mile of this station.

This record akso failed to reflect any influence attributable to the
Arvin-Tehachapi earthquake, as shown on figure 3.

Arroyo Seco near Pasadena (97). This drainage area is on the
south side of the San Gabriel Mountains. The discharge of this
stream has been measured continuously since December 1910 at an
altitude of almost 1,400 feet, at a site 5.5 miles northwest of
I'asadena.

The records failed to reflect any change in discharge attributable
to the Arvin-Tehachapi earthqualie.

Topanga Creek near Topanga Beach (98). The Topanga Creek
drainage area is located on the south side of the coastal Santa
Monica Jlountains. A continuous record of discharge of this stream
is available since January 1930, e.\ccpt for one year, at a site about
265 feet above sea level and 2 miles north of Topanga Beach.

This record does not show an.v' change in discharge resulting from
the Arvin-Tehachapi earthquake.

SUMMARY

The accelerated ground-water runoff resulting from
the Arvin-Tehachapi earthquake represents at least a
temporary depletion of the grniuid-water supplj'. In the
0.5-square-mile drainage area of ^Yons Creek in the
Santa Ynez Mountains, this accelerated ground-water
runoi¥ amounted to 2.5 inches of water over the entire
basin. The effect of this depletion could mean a reduc-
tion in runoff during subsequent years if not promptly
replaced by precipitation. Also, the effect decreases as
the size of the drainage area increases.

Evidence of the magnitude of the overall depletion
of this ground-water supply is shown on figure 19. This
diagram indicates the size of the drainage area and the
increased runoff resulting from the eartliquake in inches
of water over the entire drainage area. The volume of
ground-water runoff is for the limited 71-day period of
July 21 to September 30, 1952 for all stations except
those located in the Santa Ynez Mountains and part of
the Tecolote Investigation. In this latter instance the
volume is for the period Julv 21, 1952 through June 30,
1953.

Superimposed on the diagram is an enveloping curve
showing the maximum observed increase in runoff for
the 71-day period as a result of the earthquake. On the

Part Tl

Geology

basis of this enrvp, the volume of fn-oniid-watpr runofT mile drainage area, 0.28 inch over a 100-square-mile

amoiuiteil to the equivalent of 2.2 inches over an entire drainage area, and 0.083 inch over a 1,000-square-mile

2-square-mile drainage area, 0.94 ineh over a 10-square- drainage area.

3.0

2.0

1.0
.8

.6
.5

« •'*

" .3

Z 2

.1
.08

.06
.05
.04
.03

I ! ;

1 1 1

! ! 1 1

o Wons C

"' -C i

o Conada de

1 Refu

gio

i
Creek

1 1

0 Son J

Dse Creek

Corneros C

"\^ 1

reek

if North Fork Motilijo Creek

P\ ' ' ■ '

111 '

!^ F 1 Pn<;n C tppW -

^ >^^M.*M.;.

Crcch

■!> Tunis Creek

1 1

■^^1 i 1

^ PC

storio Cr

9ek

^ Sespe Creek nr.

X \ 'i=\<i> Caliente C
Walker Basin Creek i

Fillmore
eek

: ■ 1 1 ■ ■

k Sesoe Creek nr. Wheeler Sorinas

"'' , 1^ .. ^. . ,. 1 I'l i ;

^\^

—

Explanation

(/ Runoff from July 21 to Sept 30,1952 due to earthqi
o Runoff from July 21,1952 to June 30,1953 due to
eorthquake

ake

1 1

1 1 i

1 1

3 4 5 6 .8 1.0

3 4 5 6 8 10 20 30 40 5060 80 100 200

Drainage area in square miles

500

1000

Figure 19. Volume of runoff due to earthciuakc.

10. WATER-LEVEL FLUCTUATIONS IN WELLS

By G. H. UAVis.t G. F. Worts. jR.t and H. P. Wilson, Jr.}

ABSTRACT

Flni'ttialiiins of Rrouml-wiitcr U'\ I'l.-s cihisimI l)y the Arvin-Tfhach-
api ('arth(|uake were detected by autoiiiatit* water-level recorders
in wells as far north as Durham. Hiitte County, and as far south
as Oceanside, San Diego County. The anii)litn(le of recorded fluc-
tuations ranged from 7.84 feet in a well about 20 miles northeast
of the epicenter to 0.(112 feet in a well about 180 miles southeast
of the epicenter. Many records, especially from wells near the epi-
center, show a small residual displacement of the water level above
or lielow the level prior to the earthquake on the order of a few
hundredths to a few tenths of a foot.

Water-stirface fluctuations in wells penetratinj; unconfined aquifers
were of small amplitude, but fluctuations in nearby wells penetrat-
ing partially confined or confined aquifers were many times greater.
Although the wilter-surface movements in partially confined and
confined aquifers tend to decrease in amplitude with distance from
the epicenter, these fluctuations appear to be more directly related
to the compressibility and elasticity of the aipiifer materials than
to the degree of confinement of the aquifer or the distance of the
well from the epicenter.

INTRODUCTION

Water-level fluctuations resultinp; from earthquake
shocks have been observed for many years and have been
described in many reports (Legrjjette and Taylor, 1935;
Blanchard and Byerly, 1935; Thomas, 1940; La Rocque,
1941; Parker and Stringfield, 1950). They are of special
interest to hydrologists becau.se of the possible relation
between themag-nitude of the fluctuations and the com-
pressibility and elasticity of the water-bearinp; materials.
The Arvin-Tehachapi earthquake of July 21, 1952, is
unique because of its large magnitude, 7^ on the Richter
scale, and because of the large number of wells in diverse
types of sediments in which water-level fluctuations were
recorded. The principal damage — to buildings, oil pipe-
lines, an oil refinery at Paloma, irrigation pipelines, and
electric facilities — centered around Arvin, about 16 miles
southeast of Bakersfield and Tehachapi, about 36 miles
east-southeast of Bakersfield (Benioft', et al., 1952).

Automatic water-level recorders in California wells as
far north as Durham, Butte County, and as far south as
Oceanside, San Diego County, recorded the shock. The
rapid oscillation of the water surface appears on stand-
ard water-level charts as a vertical trace of the pen above
and below the point on the chart representing the water
surface at the time of the shock (figs. 2 and 3).

"Water-level recorders used in ground-water investiga-
tions were not designed for use as seismographs ; because
of their condensed time scale they do not record details
of the various phases of the earthquake. Blanchard and
Byerly (1935, p. 321) have shown that water-level re-
corders in wells are inferior seismographs even when
fitted with expanded-time-scale instruments because of
the damping effect caused by inertia in the system. The
fluctuation of the water surface, however, is many times
greater than comparable ground motion at the well be-
cause of hydraulic magnification. It is generally accepted
that water-level fluctuations due to earthquakes are
caused by successive dilation and compression of the
water-bearing materials, and that volume change of the
aquifer varies considerabh-, depending upon the elas-

• Publication authorized by the Director, U. S. Geological Survey,
t Geologist, U. S. Geological Survey,
i Engineer, U. S. Geological Survey.

ticitj' and compressibility of the water-bearing materials
and the earth motion in the vicinitj^ of the well. Thomas
(1940, p. 96) reported that distant earthquakes cause
water-surface fluctuations that differ in many respects
from those caused by nearby shocks. Continuity of the
trend of the liydrograph before and after the disturb-
ance and equalit\' of fluctuation above and below the
general trend line are characteristic of distant earth-
quakes, whereas fluctuations caused bj' nearby disturb-
ances commonly show more movement in one direction
thau in another, and permanent rearrangement of rock
materials as the result of the shock is sometimes indi-
cated by a change in water level. The foregoing appears
to hold true with respect to the Arvin-Tehachapi earth-
quake. Water-level records from wells near the epicenter
were characterized by inequality of fluctuations and
residual change in water level, but wells in the northern
Sacramento Valley, 300 to 400 miles from the epicenter,
showed equal fluctuation and no residual change in
water level (see table 1).

Ground water is generally thought of as existing
either under confined (artesian) conditions or uncon-
fined (water-table) conditions. Lack of confinement im-
plies free movement of water downward from the land
surface to the water surface within the containing de-
posit, whereas confinement implies lack of hydraulic con-
tinuity with the overlj-ing land surface ; that is, confin-
ing beds lie between the land surface and the aquifer
and, because of their low permeability relative to that
of the aquifer, prevent or impede vertical movement of
water. In nature perfect examples of either type are
rare. Even the least permeable aquicludes permit slow,
perhaps imperceptible, movement into or out of confined
aquifers. On the other hand, water bodies that normally
appear to represent unconfined conditions may react to
sudden stresses, such as seismic waves, in much the same
manner as confined water bodies. Presumably this eft'ect
is due to the presence of local semiconfining lenses or
layers of material of relatively low permeability which
do not prevent water-table conditions from existing on
an areal or regional scale, but which impede the move-
ment of water in response to sudden dilational or com-
pressive stresses acting upon the contaiuing deposits.
Wells should not fluctuate in response to earthquake
shocks under true unconfined conditions. Apparent
water-level fluctuations, however, may occur if the re-
corder is shaken severely enough to move relative to the
water surface, or if the shock sets up a sloshing motion
in the water in the casing, thereby causing a vertical
movement of the float. These special conditions might
exist near the epicenter of a strong earthquake but could
hardly be expected to affect wells at any great distance.

Acknowledgments. We wish to express our thanks
to our colleagues in the Geological Survej' and in other
federal, state, and local agencies who supplied helpful
advice and comments. Special thanks are given to the
following agencies which supplied records used in this
study : The California Division of Water Resources, the
U. S. Bureau of Reclamation, the Keru County Land
Company, and the City of Long Beach.

(99)

100

Earthquakes in Kern County, 1952

[Bull. 171

■a

3
>

"5 »

3
>

k
.£

3
>
3
=3
d

3
'>

a;
bO
d

£
">

_3
"d

a
£

1-
3

■S g S
1. 2 ■"

^ — 0

-a

B
d

3
>

>
_3

_3
"d

_B
d

■5 °
-0.

E a,
B

5 §

* .2

3
3

•J

B
0

01
B
CB

□

E
CB

E
IB

O
CB £

B
CB
E

■3

B
CB

-SE

"ea

B
d

-a

o o

B
0

B ^

S c

B
a^

S °

(£

PL.

J_^

•5

— 4)

1-2

—

Csj

«'

•o

„

-t*

"a.
<

■"■

'"'

■"*

„

CSI

"■J-

■v

■o

£-?

■»f

•2 2

c»

C-l

lO
CJ

o s

J3 —

^"S

bC fe

B *^

m Si

-C

a c

0.5

-T3

£
2

_a;

o fc

-C

■a

t ■£

&■"-

t--

r<.

j3 rf

f-H

r-i

■"*•

P-5

ft)

ffl

Cc;

cr;

0;
T3

-a

1
o

«"

>•

[l,

p:;

•<
>

d
3

3
03

o
>

3
>H

d
"o
Q

-D
3

!>■
"o

3
O

L-

Q
o

d
o
Q

u
I-)
.-1

•<

3
d

u
3
P3

1
1

3
E«

3
O

3
d

^ 3

c
o

o
O

"d
O

S d

u —

1
E

(0^

"d
O

•ig

B
3
O

CO
g-^

5«?

d 3

■3 O

3
O

U
a;

E d

o --

0
O

s
c

e
s

o
ft)

5
11

3
CO

E
£
■a

B
CO

-a

i|6

■a
o

■a

5
O

is
gg

41 "^

)-i

111

d 3

O
d

-a

£
o

O CD

d 3

£i

£
o

O
0

-0

00 S
Cci£

ill

■<

«c

a
a

le-; o

^ CO

•-i
CO

00
CJ

c/j

»z

■fl-

Part T]

Geology

101

*£*E

i £

S c

3 a>
a 1

E a E
.= a .2 .
> ^ > «

t >— c- ™

5 = 5 =

:? 4-

3j oi

■ - = ^ 5

•B 0

tC f-

c 2

e "^

e «3

>,r.

>i s

"aoi

1«

B 1 i

-s

■0

S 3

E| 3 S

i 0

c^ 0 C
Z "d .2 Z

^ £

confined wat
d PHoconc (?
d water in P
' (?) ailuviun'
confined wat

lli

-S s

5
"3

111

0 =

« S c 0 "a

'5'2'i

•«ti

S"isa

eS "^ «

(S D cu

Oh Cu

+ +

c V.

- 5 #-
•^ *^ o ,

•w

C»

<—

«o

>■' 1

•o

■S

1 >-

a X

>•

s >

0 '5

i 1

S >,5

■=

eIs

0 '

0
0

"cSf

C-l

y=e

^4^

St. S

o5:

I-

I a

o rn
"J »

= ^
0-3

"J i

-O o

gr si =

; = .?

C -

:^ £ d ^ ,

§2"

= ■■=

"3 5
= 05

c >

.5 — f°

I ji||2.||

« g s

r a- o -

:.■= Sc
■ ^ J:

^ = -3 ,

$^J

c 2 c o 5 ■

V^i

><
b)
>-i

•J
<

o
■J
u
(-

s a

=! 3 3

£ c = I 5 c
c^ 5 <s c ^ 0

„ c ^ c „ c

E ^ -2

>. « >.

f £ ?! S fS

= t:

= ^ p S ^ 9
^ ^ M tf == d

•^ ^^ . ^ I* 33

3 ::; eS :; cS = d
-< < < -

?•

•82

^.Z

03 o>

fe & fe s &

z z

102

Earthquakes in Kern County, 1952

[Bull. 171

a i

s i

0 a

"S 3

.2 .

at a

E i

2--

a g a g

O g

" d

■S.sl'

3 .a S "■ '

" 0) £

* S

c a, S

OJ « ^

s « i;

2 S£

q d a>
30100

t T

0) 0)

P5 •-

O .2-3

J5 a

u
o

5S
so

i &

:= O

^ o

3 § «
a oO

a" a

a ou g gu g

o23Q>o23a)oS

<: tf < ptj

r^n

■SO

■«

ffi^

•<

■<

e m

3tB

a 0

o <

u a
3 t-

S
H

.2 .9

a 9 a D

(U ■£ V Q

C o _ 2
^ .S o .9

-S-2-S'H

a s a «

a g S

j3 K j3 ^ -ja ■ 3 -a

15
■0:

fflg fflg) «< ca

0 ^ 0 ^ Q "-1 Q

B-l

■5| §'B| §'^-g §"5

38
0.

auaS-saSisa-a-sa

1-1
<

oca MOCQ MOH MO
T3 D-d atJ^T)

0
0

4)'-' Q, '-' C)^"^ 4*

(UnS ai»2 («-< tf

2 ^;

hj a td t:d

Tj" M 00 06

.-. M O* i-H

& ^ & ^

CO N M -(

\ C \ '^

52 S£ S 3

^r ^* ^r ^^

Part I]

Geology

103

»..(

■-^

■a

■§

•a

.«!

s o

P.
O

4»

CO
Li
0)

*»

■5

}..

f )

•M

-*i

« .a

^ Q

1 1

3
>

irma-

ecent

icene).

ecent

nyon

etent

a a

onfined water in La JoUa fo
tion (Eocene).

artially confined water in R(
alluvium and La Jolla fm. (Ec

artially confined water in R^
alluvium within a narrow ca
closely surrounded by comp
San Onofre breccia (Miocene

•3

nconfined water in undiff
Pleistocene alluvium.

onfined water in undiff
Pleistocene alluvium.

g
S
«
.s

-a

D U

0 0. a^

m t."

xn 10

a tT

- CO

8 S

3 +X^

d d

+ +

0 0

<. + +

CD X

0 0 r-

00 M

I '

6 d

^ ■*

ant'

o t--

« 0 00

d d

^00

O M

^ 12 S

00 o

-o

00^

3 *3

.— 4;

d d

— 41

CJ 00

0 r^ lA

2 ^

0. S

CO CO

to I-" to

elow

and

irface

efore

bock

-^ 06 0

01 S

XI — S ^ m

OD "^

00 CD

<0 •0 CD

g2 S

.S OJ j3

s b|

0.2-S

-o

. 1

1 8 gg

i i ii

o fe S

m 00 0 (O

•E s-

o a

0.-

s s

0 kC 0

0> h- 0

■o zs

N .^

M ^ N

&s|

a-o^

S J'

aj"

nT V aJ

4^ 0

T) TS -O

^ 3

'i 'i 1

o "s

rt d CO

4) 4) (U

2
O

O 3

000

^°-l"!

■0

^o "o 'o

►.-?"

Ilfi

H >•

-C

ja ja ja

s^^

>>"£ >»u >■ ti >>

3§

5d
35

Count
ilea no
Count
ilea no
Count
ilea no
Count

H
0)

S a ?.

0 a 0 s 0 s 0

•^

ut 4.2
rove, O
ut 5.3 1
gton B(

«5

U U] Ui ^
OI to V 00 oj t» 1;

5 d Q *o Q « Q

a -g a t^ a -^ a

< <

< < <

— <N

■d

< a

„,

535:

M CD Z

to to

■->

m -^

(N (N W

& gi

fe Bi ei

.y c3

o —

■*

ift iO 10

■^ -v. '^

« 1

1 1

2 § n

104

Earthquakes in Kern County, 1952

[Bull. 171

Figure 1,

Part I]

Geology

105

Figure 2. Hydrographs for six wells in the Central Valley,
July 20-22, 1952.

WATER-LEVEL FLUCTUATIONS CAUSED BY THE
ARVIN-TEHACHAPI EARTHQUAKE

Amplitude * of Fluctuations and Residual Changes in
Water Level. Water-level graphs covering the period
of the Arviu-Tehaehapi earthquake were collected from
55 wells equipped with automatic tloat-type water-stage
recorders and one well equipped with a pressure record-
ing gage. These wells are distributed from the northern
Sacramento Valley, about 370 miles north of the epi-
center, to the Twentynine Palms area, about 180 miles
southeast of the epicenter. Approximately half the wells
are in the Central Valley and the remainder are in the
Antelope Valley, the Santa Ynez River basin, the Santa
Barbara basin, the Ojai Valley, the Santa Clara River
valley, the upper Santa Ana Valley, the Twentynine
Palms area, the Los Angeles coastal plain, and the
Santa Jlargarita River valley. All the wells are in allu-
vial valleys, with the exception of those in the Santa
Ynez River basin, several of which are on alluvial up-
lands adjoining the Santa Ynez River valley. Figure 1,
a map of California, shows well numbers and locations
for all wells listed in table 1 and gives the amplitude of
water-level fluctuation for each well.

The amplitude of fluctuations ranged from 7.34 feet
in well 30S, 30E-31, near the White Wolf fault about 20
miles northeast of the epicenter, to 0.012 foot in a well

• As used here, "amplitude" is used to mean the maximum ranj^e of
fluctuation recorded by a well from hip;hest to lowest — that is,
to be equivalent to the term "double amplitude" as commonly
used in seismology and physics.

near Twentynine Palms, about 180 miles southeast of the
epicenter, in the Mojave Desert. The distribution of
fluctuations indicates that water-level fluctuations are
not wholly dependent \ipon distance from the epicenter,
but are governed also by other factors.

Residual displacement of water level above or below
the level before the earthquake appeared on many
charts, especially on records from wells near the
epicenter. Well 9N/2E-35D1, approximately 300 miles
northwest of the epicenter, was the most distant well in
which a residual change occurred. Most of the residual
changes were on the order of a few hundredths to a few
tenths of a foot,;) the three records that show residual
changes greater nian a foot are all subject to question.
Charts on wells 9N/10W-12R1 and 4S/12W-28H10 had
both slipped partlj' oif the recorder and it was not pos
sible to check the apparent displacement of water level
when the recorder chart was changed. The decline of
10.52 feet recorded in well 14S/14E-28E2 actually rep-
resented a drop in water level. The casing, however, is
perforated opposite zones of different head and the wa-
ter level in the well normally represents a compromise
between the higher head in the shallow and lower head
in the deep zone. It is assumed that the earthquake
caused enough shaking in the casing to open the well
more to water from the deep zone, thereby causing the
compromise water level to adjust downward. In the San
Joaquin Valley most of the residual changes were up-
ward, but in the Santa Clara River valley three of the

Oeplh 54 (eel

30S/26E-34B
Depth 104 feet

30S/30E-3I

Deplti 505 teet

Figure 3. HydroRraphs for five wells in the San Joaquin Val-
ley, Los Angeles coastal plain, and the upper Santa Ana Valley,
July 20-22, 1952.

106

Earthquakes in Kern County, 1952

[Bull. 171

four changes were downward. Throughout the rest of
the state residual rises occurred about twice as often as
declines, but there appears to be no significance to their
distribution. La Roeque (1941, p. 379) attributed re-
sidual water-level rise to reduction of porosity caused
by rearrangement of granular material of the aquifer
and residual decline to release of stress upon the aquifer
with some deformation of the ground-water basin.

Relation to Hydrologic Conditions and Physical Char-
acter of the Aquifers. Figures 2 and 3, which show
typical water-level fluctuations in 11 wells during the
period July 20-22, were replotted on a uniform scale
from the original recorder charts. Well 30S/26E-27C1
is believed to tap an unconfined water body, wells 1.5S/-
16E-34E1 and 19S/18E-27M1 are known to tap a eon-
fined water body, and the other seven wells shown are
believed to tap bodies of water which are \inder some
degree of confinement. Six of the records are from
paired wells — that is, wells located close together that
tap different water bodies. The paired wells are 15 -
16E-34E1 and 15S/16E-20R1, 19S/18E-27M1 and 19S/-
18E-27X1, and 30S/26E-27C1 and 30S/26E-34B1.

Water-level fluctuations in the wells penetrating un-
confined aquifers are of small amplitude. For example,
wells 30S/26E-27C1 and 30S/27E-5H1, 54 and 50 feet
deep, respectively-, which both tap unconsolidated sandy
alluvium of the Kern River alluvial fan, showed an
amplitude of only 0.35 and 0.20 foot, respectively, in
spite of the fact that the wells are within 30 miles of the
epicenter.

Consideration of the fluctuations in wells known to
penetrate partially confined and completely confined wa-
ter bodies suggests that the degree of confinement does
not necessarily control the amplitude of water-level fluc-
tuation. For example, at paired wells 15S /'16E-34E1 and
15S/16E-20R1, which tap a confined aquifer and a par-
tiallj' confined aquifer, respectively, the fluctuation in
the partially confined water body was considerably
larger than that in the confined water bodv. Converselv,
at paired wells 19S 18E-27X1 and 19S''18E-27M1, which
tap the same partially confined and confined water
bodies as wells 15S/16E-20R1 and -34E1, the fluctuation
in the well tapping the confined water body was 10 times
greater than that in the well tapping the partially con-
fined water body. The fluctuation in well 13S/17E-10A,
which is believed to tap partially confined water, was

considerably greater than the water-level change in
either well 15S/16E-34E1 or well 19S/18E-27M1, both
of which tap an aquifer confined beneath a thick later-
ally extensive clay bed which underlies most of the
western San Joaquin Valley. Well 30S/30E-31, which
registered an amplitude of 7.34 feet, is believed to pene-
trate only partially confined water bodies. Several wells
in Kern County reported to penetrate partially confined
aquifers showed water-level surges in excess of 5 feet,
which was the limit of fluctuation the particular instru-
ments could record.

The records presented from the Arvin-Tehachapi
earthquake suggest that water-level fluctuations are
more directly related to the compressibility and elas-
ticity of the aquifers than to the degree of confinement
of the aquifers or even to distance from the epicenter.
The compressibility and elasticity of aquifers, in turn,
appear to be related to the lithologie features of the
materials. Studies by foundation engineers (Terzaghi
and Peek, 1948, pp. 57-61 ) have shown that fine-grained
materials such as clay and silt are far more compressible
than sand. However, because the water level fluctuations
in question represent an instantaneous response to
stresses on the aquifer, it seems unlikely that thick clay
or silt lenses within an aquifer would yield much water,
because of their low permeabilit.v ; hence they would
have little effect on the water-level fluctuation. Jacob
(1941, p. 577) indicates, however, that if a sufficient
number of clay laminae are interbedded with sand in an
aquifer, the release of stored water from the clay is
virtually instantaneous and the quantity of water yielded
is related to the modulus of compression of the clay and
the thickness, configuration, and distribution of inter-
calated claj' beds. The mineralogy of the constituent
sands also may exert an important effect upon the com-
pressibility of an aquifer. Terzaghi (and Peck, 1948, pp.
57-60) has demonstrated that a mica content in excess
of 10 percent can raise the compressibility of a sand to
that of a soft clay. Alluvial fill in California basins,
especially that derived from granitic source areas such
as the Sierra Nevada, the San Gabriel and San Ber-
nardino Mountains, and the Peninsular Range, might be
expected to have an appreciable mica content, on the
order of 10 percent or more. Hence, it is likelj' that this
factor may influence the compressibility of the aquifers
tapped by many of the wells considered herein.

11. SEISMIC PROSPECTING FOR PETROLEUM AND NATURAL GAS
IN THE GREAT VALLEY OF CALIFORNIA

By Joshua L. Soske

ABSTRACT

Tho refraction seismoRrapb was intrcKhioi-d as a niPthod of
invpstiKatinK geologic structure in the (Jreat Valley of California
enrly i]i 1926. After several years of exhaustive testing it was
found to be inadequate to solve the problems of the San Joaquin
Valley and was discarded in 1029 as a routine method of study.
Tlie reflection seismograph was introduced for the study of the
buried structure in the Delano-Rakersfield area by the Geophys-
ical Kesearch Corporation in 102S and became the generally
accepted method for oil and gas prospecting in the areas of no
rock outcrops. The success of the reflection method in making
locations for the discovery tests of Huena Vista Lake and Chow-
chilla gas fields in 1934 stimulated experimental work resulting in
improvements applicable to the problems of the Great Valley. The
limited applicability of correlation methods of reflection study
led to the development and wide use of the dip method of analysis
and the construction of maps of phantom horizons to represent
the buried geologic structure. As a result of continued improve-
ments in seismic instruments and techniques, many areas were
rcsurvpyed from time to time. The success of the reflection seis-
mograph in the Great Valley is indicated by the di.scovery of 34
oil and gas pools in which the method contributed essential tech-
nical information in advance of drilling. If new oil and gas pools
are to be found in the future at the rate established during the
last 20 years, new geological approaches and technical improve-
ments in suitable geophysical methods seem necessary.

INTRODUCTION

The seismograph was introduced in California in its
early stages of development to aid in the search for oil
and gas. Only in the Gulf Coast area and Oklahoma was
the method used earlier. No attempt will be made here to
discuss the technical aspects of the method involving
the theory of elasticity, wave propagation or computa-
tions, but an attempt will be made to present a rather
broad picture of the principles of the seismic method
and how it has been applied in the search for gas and
petroleum.

Oil and gas pools exist in various geological settings.
Several prerequisites are necessary: (1) Sources for the
generation of petroleum or natural gas. These are usually
recognized as dark-colored carbonaceous shale beds. (2) A
reservoir rock with sufficient porosity and permeability
to allow for the accumulation and withdrawal of the oil
and gas in sufficient quantities to be commercial. This
rock is usually sandstone, fractured shale, or porous
limestone. (.3) A suitable geological trap to facilitate
the gathering and storing of the oil and gas underground
in sufficient volume to justify commercial extraction.
Tj^pes of these structural traps include closed anticlines,
domes, faulted anticlines, noses, wedgeouts, and litho-
logical traps.

The problem of whether source beds and reservoir
rocks are present is principally one for the geologist.
The prospecting geophysieist using seismic methods is
interested in mapping the underground geological struc-
ture in hopes of finding favorable oil or gas traps.
The oil operator is also interested to a lesser degree in
learning about unfavorable areas, i.e., synclinal areas.

Where geological exposures of the formations reveal
the structure of the area the services of the geophysieist
are not required by the geologist. But in those eases
where the exposures are hidden from the view of the
field geologist, he often needs the services of the geo-

physieist to map the structures of the buried rocks. Much
of the Great Valley of California presents this problem
of geology covered by alluvial deposits. To solve the
problem the geophysieist often uses the refraction method
to study the characteristics of the overburden and the
reflection shooting method to determine the structure
of the buried sedimentary rocks. The seismic method is
not one that locates oil or gas directly. It is a method
that investigates the existing geological conditions to
determine whether they are favorable or unfavorable
for the possible commercial production of oil or gas.

Acknowledgments. The writer wishes to acknowledge
his indebtedness to many friends who have reported first
hand experiences with seismic prospecting in California.
If the writer could recall all their names, the list would
be too long to include here. It is a pleasure to acknowl-
edge the assistance of a number of individuals who sup-
plied material and valuable suggestions used by the
writer in the preparation of this paper.

Mr. Henrv Salvatori furnished information concern-
ing the early .seismic work, Mr. Milton C. Born supplied
data regarding early work of the Geophysical Research
Corporation in California, Dr. H. B. Peacock supplied
reproductions of early sei.smograms recorded in the San
Joaquin Valley, Dr. Frank E. Vaughan supplied photo-
graphs and descriptions of early refraction work in Cali-
fornia, Messrs. "W. D. Goold and Downs McCloskey re-
ported on Frank Rieber's early refraction experiments
and supplied photographs, Mr. Carl H. Savit supplied
photographs of Western Geophysical Company's early
equipment and E. Fred Davis collaborated by supplying
factual information on early refraction shooting in Cali-
fornia.

ARTIFICIAL EARTHQUAKES

Both the refraction and reflection methods of seismic
prospecting make use of artificial earthquakes. Charges
of dynamite are placed in holes in the ground and then
detonated, causing the ground to be agitated by the quick
and sharp impact of the blast. This initiates the elastic
waves which are propagated through the subsoil in
every direction.

The terms refraction and reflection have been bor-
rowed from optics and are used to describe the path
of elastic waves in much the same manner that these
terms are used to explain the wave paths of light rays.
Variations of the velocity of light in substances cause
refraction and reflection and similarly such variations
in seismic wave velocities cause similar events to take
place along the paths of the elastic waves.

Seismic waves obey Snell's law, Fermat's and Huy-
gen's principles. The basic law of refraction, usually
referred to as Snell's law, is actually a consequence of
Fermat's principle. This law states that the ray follows
the fastest path in traveling from a given point to a
second point in a mediiim. If the medium through which
the ray passes provides a constant velocity, then the
path is a straight line; if, however, the second point on
the path is in a second medium providing a velocity

(107)

108

Earthquakes in- Kerx County, 1952

[Bull. 171

different from the first, then the path is not a straight
line and the bending of the path takes place at the
point or points where the velocity changes.

THE REFRACTION SEISMOGRAPH METHOD

In using the refraction method, the bending of the
ray path makes possible the penetration of the earth's
crust by the seismic wave at one point and its reappear-
ance back at the surface at a second point where it may
be observed with seismographic instruments. The refrac-
tion method is based on the condition that if we can
measure the seismic velocities, that is, the speed of an
impulse through the various layers of the subsoil, then
we can graphically plot the ray path in space.

As a general rule when high explosives are detonated
in the ground the forces released in the rock far exceed
the strength of the rock-forming materials. This leads to
a zone of complete failure which extends outward in
all directions from the center of a well-confined blast.
At greater and greater distances from this destructive
center the effective force on the rock decreases. Inspec-
tion of the results of a blast reveals that at the origin the
rock is violently ruptured and crushed but the evidence
of this great physical force decreases rapidly as the dis-
tance from the explosive center is increased. In brittle
rock we find a more or less spherical zone of shattered
materials completely enclosing the central space of total
destruction.

Examination beyond the severely shattered zone dem-
onstrates that the degree of shattering decreases until
finally there is a gradation into rocks that have not
suffered any visible rupture or permanent set as a result
of the blast. In this zone the strength of the rocks ex-
ceeded the applied effective forces, and the reaction of
the rock to the quick and sliarp compressional forces was
in the form of elastic deformation or strain. Here the
compressional pulse wave is initiated by the elastic re-
sponse of the formations. The initially disturbing pulse
wave is propagated outward from the zone as a spherical
wave front. The oscillatory nature of the origin maj- in

FiGUKE 1. A refraction wave-front diagram sketched for a
three-layer problem .showing the effect of a change in slope of the
buried surface of the third layer. The magnitudes of the seismic
velocities were chosen to approximate those experienced where the
uppermost layer consists of dry alluvial gravels, sands and clays,
the intermediate layer the same as the uppermost except saturated
with water, and the third and deepest layer of crystalline bed rock.

part be accounted for by the release of strain energy,
which was momentarily stored during the very short
period of the initial deformation caused by the sudden
application of the compressional forces and transferred
to the zone of no failure. The nearly unrestrained relief
of this deformation of the rock causes a rarefaction ef-
fect to follow the release of compressional deformation.
This disturbance is likewise propagated as a pulse wave
front following the first. The rarefaction may exceed
equilibrium conditions and thus initiate a second but
less" vigorous compressional pulse at the origin. In this
manner the particle oscillation at the shot point is rap-
idly reduced to an insignificant amplitude thus giving
rise to a short train of damped wave pulses.

Ware Front Diagram. Figure 1 is called a wave
front diagram and represents successive positions of the
traveling pulse front plotted at equal time intervals (10
milliseconds) following the instant of the origin. It il-
lustrates how the ray paths and wave fronts propagate
from the center of the blast space to deeper aiul higher
speed layers, thence along the surface contacts of these
layers and back to the ground surface. As the ray paths
are always normal to the wave fronts this scale drawing
of a vertical section through the wave fronts shows how
the ray taking the detoured path through the higher
speed beds may arrive at a point at the surface simul-
taneously with the pulse which takes a more direct route
at a lower velocit.v through the surface layer. This point
where the rays arrive simultaneously marks the so-called
"critical distance" (xc) from the shot point. Beyond
this point the first arriving rays are those that liave
been refracted and thus indicate the presence of the
higher speed layers at depth.

Above the vertical section the travel-time curve has
been plotted. This graph represents the data measured
in the field. The reciprocal of the slope of this curve
is the apparent velocity of tlie seismic wave along the
ground surface. When the boundaries of the layers are
horizontal tlie apparent surface velocity is equal in mag-
nitude to tlie actual velocity of the deeper layer pene-
trated by the observed wave. When the apparent veloci-
ties deviate from the true seismic wave velocities of the
various layers, the geophysicist is given a clue to the
attitude of the layers. This is illustrated by the portion
of the travel-time curve showing the most distant ar-
rivals of the refracted waves from the inclined boundary
of the third layer of the diagram.

In this manner of making measurements of the appar-
ent surface velocities the prospector can often obtain the
depth to the various layers, tlie true seismic velocity of
the various layers and in many places he can determine
the attitude of the layers. The determination of the seis-
mic velocity of the buried media gives a clue as to the
physical properties of the subsurface materials because
the seismic velocity is a consequence of the physical
characters of the rock, i.e., density, bulk modulus and
rigidity. Water saturated gravels, various types of bed-
rocks, rock salt, sedimentary and igneous formations
may be located and identified by this method of studying
seismic travel-time data.

Despite the tremendous success attained (40 salt
domes were found by this method 1924-32) through the
use of the refraction metliod and the large number of

Part II

Geology

109

Figure 2. A 24-trace seismogram recorded in the San Joaquin
Valley illustrating the arrivals of direct, refracted, and reflected
pulses. Split spread spacing 1200 feet, 400-foot gap, 800 feet, no
offset. Courtesy Western Geophysical Company.

prospects surveyed, such as dam sites, bedrock, and salt
dome, only a few seismoo-ranis and travel-time curves
have been published. Now tliat the refraction method
has been largely superseded by the extensive employ-
ment of the reflection seismograph there is little hope
that man}' of these will ever be published.

THE REFLECTION SEISMOGRAPH
In general, reflection seismograph prospecting re-
quires much more elaborate instrumental apparatus and
more field equipment, such as power driven shot hole
drills and water trucks, than refraction prospecting.
This wide difference in equipment stems from the fact
that the two methods are concerned with the observa-
tions of pulses of very unlike characteristics. In reflec-
tion shooting we employ an echo method and, therefore,
we are especially concerned with the recording of late
arriving events on the seismogram. The time required
for the round trip of the wave from the shot point down
to the reflecting horizon and back to the surface is in
general much greater than the time necessary for the
direct or refracted wave to travel only one way from
the shot point to the recording seismometer. The arrivals
of the direct and refracted waves are usually the first
events recorded on the seismogram. In refraction work
nearly all the interpretation may be based on these
first impidses recorded, often referred to as the "first
breaks." In refraction studies only one seismometer may
be used and the refracted pulses may be readily identi-
fied on a single trace seismogram. On the other hand,
reflected pulses recorded on a single trace seismogram
would be indeed difficult to identify. Several single trace
seismograms involving different distances from the shot
point to the recording seismometers would be necessary
to identify a certain reflected pulse. One good reason
for the use of multitrace seismograms in reflection seis-
mograph work is that the later pulses recorded on the
seismogram after the first breaks may be either direct,
diffracted, refracted or reflected pulses. In general prac-
tice the reflections are identified on the seismogram by
the distinct pattern in which the reflected pulses appear
on a multitrace seismogram. The pattern is frequently
referred to as the "line up." Early reflection equipment
employed onl\- four or six recording traces while current
reflection seismographs employ 2-4 to 48 traces. Each
trace represents a separate channel of recording eon-
sLsting of three units, one or more seismometers, an
amplifier and a recording galvanometer.

The patterns of the reflected pulses on the multitrace
seismogram are distinctive because all of the reflected
pulses, from a nearly horizontal reflecting interface, ar-
rive at the ground surface almost simultaneously and
in like phase, thus similarly activating all the seismom-
eters almost at the same instant. Figure 2 is a repro-
duced reflection seismogram showing first break data
and marked reflections. The fairly good line-ups of
recorded pulses of similar characters are recognized by
(a) similar wave form, (b) similar wave length, (c)
similar groups of wavelets, (d) similar amplitudes. These
criteria have been used to identify the reflections as
marked on the reproduced seismogram of figure 2. The
reflected wave fronts which arrive at the seismometers
are usually much more nearly horizontal than are those
representing refracted waves.

110

Earthquakes in Kern County, 1952

[Bull. 171

There are many specifications which must be met in
the design of reflection recording equipment. Some of
the more important of these are (a) great timing ac-
curacy (one part in ten thousand), (b) high damping
throughout the recording system, (c) suitable ranges of
frequency responses (filter circuits), (d) an adequate
method of automatic volume control, (e) high sensitiv-
ity, and (g) multiple seismometer arrangements. Only
great timing accuracy and high sensitivity are critical
specifications for refraction equipment.

The field procedure for the two methods is similar
in that both require that the explosive charges be placed
in that portion of the ground that is characterized by
efficient elastic transmission characteristics. One excep-
tion to this statement is the procedure making use of
explosions in the air for recording of reflections. When
shooting in the ground the charges should be well-con-
fined so as to avoid the transfer of the blast energy to
the atmosphere. Water is usually used as the stemming
material. Explosive charges range from half a pound
to 5 pounds in routine reflection shooting; however,
slightly larger charges are sometimes used. The charges
are always detonated by electric type blasting caps.

Briefly, the function of seismic equipment is to amplify
and make permanent recordings of the feeble displace-
ments of the ground and to eliminate or decrease the
influence of undesired impulses, i.e., wind, rain and
other background noises, as much as possible. In prac-
tice it has been found that apparatus which is capable
of a maximum magnification of about ten million (about
ten millimeters displacement of the trace on the seis-
mogram corresponding to one millimicron displacement
of the ground) is satisfactory.

In seismic work only two items are measured, time
and distance. The disposition of the configuration of tlie
seismometers is measured in feet and the elapse of time
between successive events is measured in seconds. The
distances are a matter of plane surveying in the field
and the elapse of time is taken from the seismogram. If
the round trip time for a certain reflection is obtained
from the seismogram, then the distance traveled by the
wave during this time is obtained as the product of the
average velocity and the observed time. This gives the
round trip distance. The depth is approximately half
this distance. In precise work needed for detailed
geological structure, adjustments are frequently made
for the curved paths of the rays due to refraction and
variations in seismic velocity. For example, if reflections
traveling through the Tertiary sediments of the Great
Valley of California, where the dip of the bedding planes
is approximately horizontal, gave observed reflection
times of 0.500, 1.000, 2.000, 3.000 and 4.000 seconds, then
the reflections arise respectively from interfaces 1,530,
3,350, 8,020, 14,250 and 21,700 feet below the surface of
the ground.

The reflection method is ideally suited to mapping of
the attitudes of deeply buried sedimentary rocks. The
effect of a dip as small as 50 feet per mile can be detected
under favorable conditions. The various layers of shale,
sandstone and limestone give good reflections as a gen-
eral rule over most of the floor of the Great Valley of
California. Some difficulties have been encountered in
areas where the seismic wave transmitting qualities of
the overburden are either too good or too poor. A portion

of the southernmost area of the San Joaquin Valley is a
place in which the rather thick overburden is a poor
transmitter of seismic waves and it is difficult to obtain
good reflections consistently in this area. A few small
areas in the Great Valley have overburden that is too
good in its transmitting quality for the recording of good
reflections. These handicaps are now being overcome in
some degree by various patterns for the positions of
seismometers, multiple seismometer arrangements, multi-
ple shot holes, air shooting and new instrument designs.

THE NEW ERA IN PETROLEUM PROSPECTING

The First Oil Pool Found by Geophysical Methods.
A new era in petroleum prospecting was ushered in when
geophysical methods were credited with the discovery of
the Nash Salt Dome, Fort Bend County, Texas in 1924
by the Rycade Oil Company. Actual oil production on
the flank of the dome was not attained until 2 years
later but the presence of the salt dome was confirmed by
the very first test well. Long Point Salt Dome was also
discovered the same year by the Gulf Oil Company mak-
ing 1924 an important date in the history of geophysical
prospecting for petroleum. These first successful investi-
gations involved gravity studies and definitely demon-
strated that conditions favorable for the subsurface
occurrence of crude oil could be located deep within the
earth's crust by means of instruments at the surface of
the ground. The attention of the early oil prospectors
immediately turned to reviewing the available knowledge
on all natural phenomena that might lead to new geo-
physical methods.

There can be little doubt that this review led to a
thorough examination of fundamentals and principles
that had been applied by seismologists in their studies of
earthquakes and large accidental explosions. These early
studies indicated that information on the surface forma-
tions could be obtained by measuring the velocities,
frequencies and energies of the seismic waves propagated
through the earth's crust. The application of the princi-
ples used and the information gained by the earthquake
seismologist led to the development of the seismic method
of geophysical prospecting.

Acceptance of the Refraction Method. In 1924 the
refraction seismographic method of prospecting was ac-
cepted as one of the important tools of the exploration
geologist, a result of the discovery of the Orchard Salt
Dome in Fort Bend County, Texas. The results obtained
were possible because of the relatively uniformly low
seismic velocity of the near surface rocks which was
easily distinguished from the higher velocities of the salt
or associated cap rock of the salt domes. This led to a
rapid succession of oil pool discoveries in this type of
geologic structure. During the following 8 years the re-
fraction seismograph enjoyed a high place among the
tools used by the oil prospectors in 'Texas and Louisiana.
In this period the refraction method was credited with
the finding of 40 salt domes.

Refraction Method Introduced in California. Only 2
years after the seismograph had been applied to geologi-
cal problems in the Gulf Coast area it was tested in
California. Mr. W. D. Goold of the Tide Water Associ-
ated Oil Company reports that he was a member of a
refraction seismograph crew which began field tests west

Part I]

Geology

111

Figure 3. Fnink Kielier recoidiiif; truck and crew operating
in the San Joaquin Valley during 1926. I'huto supplied by Douns
ilcClosky.

of the Lost Hills oil field in June 1926. The experimental
work was performed by Frank Rieber and jointly sup-
ported by the Standard Oil Company of California, the
Associated Oil Company, and the General Petroleum
Corporation.

The equipment used by the Rieber organization em-
ployed a new idea in seismometers in usiuij the piezo-
electric property of a ([uartz crystal. In use the crystal
was loaded with a relatively large inertia reactor in such
a way that the vertical motion of the earth imparted an
acceleration of the inert mass which exerted varying
pressures on the confined crystal. This pressure variation
on the crystal caused electrical potential differences be-
tween opposite faces of the crystal. These potential dif-
ferences were amplified b.y means of an electronic
amplifier and then applied to a galvanometer, whose re-
action was recorded on a moving strip of photographic
film resulting in the seismogram.

Advantages claimed for this new seismometer were (a)
that the seismometer contained no moving parts such as
masses supported by springs, (b) that the generated
voltages are proportional to the acceleration of the
ground, whereas the outputs of most other seismometers
were proportional to the velocity of the ground motion.
The fragility of the crystals and the very low sensitivity
of the Rieber refraction seismometer may have contrib-
uted to the general failure of these earl.v tests. Relativelj^
large charges of explosives were iised. Goold describes
one of the experimental tests as follows: "The single
seismometer was placed at a distance of 1 mile from the
shot point, which consisted of 30 holes 4 inches in diam-
eter, drilled by hand into the earth to a depth of approxi-
mately 6 feet. A total charge of 1,500 half-pound sticks
of 80 percent strength gelatin was distributed among
the holes so as to fill them about half full with dynamite.
All the holes were detonated simultaneously with the use
of 'Cordeau' fuse. This resulted in a terrific explosion, a
tremendous dust cloud and sometimes a seismogram."

Rieber 's refraction crew made further tests in the
vicinity of the present Coalinga Nose oil field during
the early part of 1927. It is reported that this work
demonstrated that the water table at a depth of about

300 feet could be successfully mapped as a high-speed
marker layer. It seems that no hint of the presence of
the Coalinga Nose Pool was obtained by this work.
P''igure 3 is a photograph of Rieber 's seismogram re-
cording truck and some of the crew.

In 1930, Frank Rieber published a report on some of
his early experiments with the refraction method in
which he pointed out the effects of the unconsolidated
nature of the near surface formations on the refraction
travel-time curves and recognized the increase of seis-
mic velocity with increased depth of the sedimentary
formations of California. In the opinion of the writer
the results of Rieber 's work would have been much more
important had there been any possibility of finding salt
domes in California.

FiGtrRE 4. Slicll Oil Company refraction seismograph recordinpr
crews operating in the San .loaquin Valley in 1928. Photo supplied
hy Frank K. Vttuyhan.

Shell Oil Cvmpany Tests Refraction Methods. Fol-
lowing Rieber 's experiments probably the next attempt
to test the refraction method as a tool to assist in the
study of buried geologic structure of the San Joaquin
Valley was made by the Shell Oil Company. Shell began
these tests in 1927, contiiuied the work through 1928
and 1929. Dr. Frank E. Vaughan provided the general
historical sketch of these early operations in California.
Scliweydar meclianical two - component seismographs
were employed in these tests. Two complete seismometer
mechanisms were built into a single instrument. In
general the two systems consisted of two inertia masses
mounted on leaf type springs in such a manner as to
respond to vertical and horizontal components of the
earth movements. The movement of each mass relative
to the frame of the seismometer was first magnified by
a long, light, stiff lever attached to the inertia mass.
At the end of each lever a bow was attached which
usually carried a human hair as a bow string wound
around a slender spindle to which a small plane mirror
was attached. Any slight movement of the bow caused
a corresponding rotation of the mirror which reflected
a beam of light that fell on a cylindrical lens at a
distance of 1 meter. The beam of light was brought to
a focus by the lens on a moving strip of photographic
paper. The photographic paper was contained in a sep-

112

Earthquakes in Kern County, 1952

[Bull. 171

afi I

Figure 5. Ailjusting meclu-inical seismograph, Shell Oil Company,
l!t28. Photo supplied by Frank E. Vaughan.

Ill spite of the preneral opinion tliat the results of the
early refraction tests were conjectural, there is some
evidence that the results of the Shell Oil Company work
were useful to the petroleum prospector of that period.
This work aided the Shell Company in the following
ways: (1) The Shell Oil Company was apparently able
to increase their knowledge of the general structure of
the San Joaquin Valley by more mapping the basement
complex along the eastern area as far out into the valley
as Hauford. (2) Their refraction work aiforded addi-
tional proof that the gravity anomalies detected by early
torsion balance surveys along the eastern portion of the
San Joaquin Valley were probably due to variations in
density of the buried complex rocks, rather than to
deformation of the basement surface or the overlying
sediments. This idea was suggested by an earlier study
of the comparison of magnetometer and torsion balance
surveys. (3) There seems to be little ((uestion that this
work supplied the Shell Oil Company with some infor-
mation on local geologic structure in the southeastern
portion of the valley. Moreover, if this rather expensive
and hazardous procedure had not -given some usable
information the tests would not have been continued

arate instrument called the camera which was driven
by a clockwork type of mechanism. It is reported that
only the registration of the vertical ground motion was
of any actual value. The frequency of these instruments
was about 15 cycles per second ; the magnification of the
ground motion, apart from any possible slipping be-
tween the bowstring and the spindle, has been estimated
to range between 14,000 and 40,000. Little or no damp-
ing of the seismometer mechanism was provided in this
work because a record of the first arrivals of the waves
only were desired. Charges of explosives used in these
studies ranged from 300 to 600 pounds of 60 percent
nitro-gelatin powder; however, a few charges as large
as 1200 pounds were used. Experiments with 10 percent
ammonia powders are reported to have given good re-
sults in the absence of water.

Figure 6. SeismoloRist contMctiiiR shooter by radio, Shell Oil Com-
pany, 1928. Photo supplied hy Frank E. Vaughnn.

Figure 7. Mushroom shaped refraction blast, San Joaquin Valley,
Shell Oil Company, 11(28. Photo supplied hy Frank E. Vaughan.

for almost 2 years. In general it probably indicated dis-
placements of faults in the shallow basement complex
and may have suggested the structure in the overh-ing
sediments as a consequence of the faulting.

Very little refraction work has been done in Cali-
fornia since 1929 except for a few very special types
of surveys. During 1937 and 1938 the Geophysical Engi-
neering Corporation mapped the basement complex sur-
face in the vicinity of Arvin, California using the re-
fraction method. The problem of seismic wave penetration
to the basement rocks was solved by the innovation of
placing the shot point directly in the outcrop of the
high-speed basement rocks. The seismometer spreads
were placed along radial lines containing the shot point
as a means of observing the travel-times for the wave
to pass along the basement interface and uji through the
overlaying sedimentary formations to the surface. Dif-
ferences in observed travel-times were attributed to

Part I]

Geology

113

eitlier different horizontal distances fi'oni the shot point
or different thieknesses of the sediments. When the hori-
zontal tlistanees were equal between the sliot point and
different recording; stations then the dilferenees in the
travel-times were attributed to variations in the depth
to the basement complex. This involves the reasonable
assumption that the velocity of the seismic wave is
nearly constant in the basement rocks over the distances
recorded, which in this case varied from a few thousand
feet to 8 and 10 miles from the sliot point. In this man-
ner it was a relatively easy matter to make a map of
the surface of the basement rocks at depths in excess
of 1 mile. The general purpose of the survey was to
locate possible buried scarps, ridges and valleys in the
basement rock surface of the area.

No discussion of early seismic work in California can
ignore the activities of the Geophysical Research Corpo-
ration, a subsidiary of Amerada Petroleum Corporation.
Mr. M. C. Born, Geophysical Supervisor for the Ame-
rada Petroleum Corporation in California reports the
following early activities of his company. Dr. 11. B.
Peacock was in charge of a Geophysical Research Cor-
poration crew which operated in California during 1928

Figure 0. L.Trge refraction blast near a San Joaquin Valley
orchard, Shell Oil Company, 1928. Photo supplied hy Frank E.
Vaughan.

First Seismic Reflections Ohserved hi California by
the Geophysical Research Corporation in 1928. There
seems to be no doubt about the fact that the reflections
recorded in 1928 were the first to be observed in Cali-
fornia. Figure 13 is a photograph of one of GRC 's early
refraction units which operated in California. Figure 14
is a photographic reproduction of two California seismo-
grams recorded by GRC in 1928 illustrating refraction
pulse arrivals, reflections, ground roll, and blast-phone
breaks. Tlie last was used to compute the distance be-
tween the shot point and the receptor.

The Geophysical Research Corporation terminated
their seismic work in California in 1929 but returned

Figure S. Dry rotary shot hole drill. Shell Oil Company, 1928.
Photo supplied by Frank E. Vaughan.

and 1929 for a period of 21 months. This crew was oper-
ated jointly for the Amerada and the General Petro-
leum Corporations. The work was done in the Bakers-
field-Delano area. Initial work comprised a refraction
shooting program using three recording trucks, each
equipped with a single channel recording system with
radio code communication and a blast-phone detector for
the purpose of measuring distance by using the velocity
of the sound of the blast through the air. A dry rotary
drill, similar to that used for present day telephone hole
placements, drilled the shot holes to depths of about 15
feet. Charges up to 500 pounds of explosives were det-
onated. Analysis of these early records disclosed the
basement complex reflection which in some places could
be correlated over considerable distances. With this en-
couragement, a four-channel recording system was in-
stalled in a truck and one crew operated as a reflection
crew.

•*>)•

:^Ji

'iy-

Figure 10. RecordinK car and tent enclosing seismograph, Shell
Oil Company, 1928. Photo supplied by Frank E. Vaughan.

114

Earthquakes in Kern County, 1952

[Ball. 171

FiGlTBE 11. Large refraction shot, San Joaquin Valley, Shell Oil
Company, 1929. Photo supplied by Frank E. Vaughan.

early in 1932 with improved equipment. The first unit
was soon joined by several others in the same year and
this group began the routine seismic work which has been
continued by the Amerada Oil Corporation in Califor-
nia up to the present time.

In 1925, the Seismos Company had four refraction
crews operating in the Gulf Coast area, and was doing
practically all of the commercial seismic work in the
United States. The Seismos crews employed mechanical
seismographs which were relativel.y low in sensitivity and
efficiency. That same year, the Geophysical Research Cor-
poration was organized, in the hope of improving seis-
mograph technique and thereby widening the use of the
instrument. GRC's first achievement was to introduce
an effective electrical seismograph with a much improved
sensitivity and to employ radio communications between
the recording and shot point stations. The first GRC
Refraction Crew was placed in the field under the direc-
tion of Dr. E. E. Rosaire in the Spring of 1926. At this

early date the seismograph as a tool for exploration was
in its infancy and the radio art had not progressed very
far. The vacuum tubes of this period were not very eflS-
cient and it was not an easy problem to build a portable
and stable audio type amplifier. By the end of 1926
GRC had five or six refraction crews working in the
Gulf jCoast area and one experimental crew testing the
possibility of recording reflections. Messrs. J. E. Duncan
and Heury Salvatori worked together on this GRC re-
search project during December 1926 and by March 1927
they had succeeeded in recording and identifying their
first seismic reflections. This seems to have been the very
beginning of the technique of using the reflection method.
During the summer of 1927, GRC, in addition to
maintaining their large number of refraction crews, in-
creased the number of seismic reflection crews to four,
and systematically prospected the Seminole Plateau of
Oklahoma. It seems that the reflection method was not
used commercially by any company other than GRC
until 1929. The status of the now extensively used reflec-
tion shooting method is indicated by the remark of Don-
ald Barton in his report before meeting of the Amer-

FlGURE 12. Dry n.iaiy sh..| lioli- ilnii sliowing details of machine,
Shell Oil Company, 192S. Photo supplied by Frank E. Vaughan.

Figure 13. Early reflection and refraction recording equipment
used by Geophysical Research Corporation near Bakersfield, Cali-
fornia, 1928. Photo supplied by II. B. Peacock.

ican Association of Petroleum Geologists in 1929. He
stated, "the reflection shooting being tested by the
Geophysical Research Corporation has not been com-
pletely accepted as an acceptable method, though GRC
seems to have confidence in it ! "

With this experience, it is not surprising to find that
GRC tested the reflection seismograph in California as
early as 1928. However, it was not until the years 1931
and 1932 that the reflection method was persistently
tried in California by other operators. The first year
of this work was somewhat disappointing. The areas
chosen for these initial tests were, on reworking with
modern equipment, found to be rather unfavorable lands
for the application of the reflection seismograph. Never-
theless the tests gave some slight encouragement for
further testing of the method.

This minor encouragement was not considered suffi-
cient to justify the supporting reflection seismograph
crew by a single oil company. Consequently the opera-

Part II

Geology

115

* X .

tion of a Western Geophysical Company seismop;raph
crew was supported jointly by several oil interests on
a more or less experimental basis. The area chosen for
this joint seismic study was in the vicinity of Merced,
California, where the ground conditions were found to
be more favorable to the problem of obtaining reflection
seismograms. Almost from the very beginning of this
joint operation, good reflections were obtained. At an
early stage of the work the efforts of the pioneer reflec-
tion seismologists were rewarded with two and as many
as four good reflections on a large percentage of the
seismograms. It is reported that at least one of these
reflections could be correlated over much of the area.

The seismographic results indicated a geologic struc-
ture about 16 miles south of the city of Merced. Pure
Oil Company tested it with a well completed in Novem-
ber 1934 as the discovery well of the Chowchilla gas
field. The principal interest in this discovery is probably
an academic one : that is, it marks one of the very early
natural gas discoveries as a result of systematic reflec-
tion seismograph work in California. As a natural gas
producer the Chowchilla gas field has not been very
important because of the low heating value of its gas.

There is some evidence that the first natural gas field
discovered in California by the aid of the reflection seis-
mograpli was the Buena Vista gas field brought in by the
Ohio Oil Company on July 11, 1934. It was completed
four months in advance of the Chowchilla gas field dis-
covery.

Amerada 's discovery of the Tracy gas field in 1935
may be considered the second discovery of a commercial
gas field in California found by the use of the reflection
seismograph. This field is of importance because it was
the first commercial gas field discovery to indicate that
the 'northern portion of the Great Valley of California
contained natural gas accumulations in significant
quantities to interest the prospector. The geologic struc-
ture of this field is that of an elongated dome covering
an area of about 600 acres. No hint of this field is dis-
cernible from the surface and no subsurface information
from wells drilled in the area was available prior to the
discovery. It is therefore concluded that this discovery
was the direct result of reflection seismograph work.

Figure 14. Reproduction of early seismograms recorfled by
Geophysical Research Corporation, 1928, near Delano, California.
Shows timing trace, arrivals of refractions, reflection, ground roll,
and air sound wave. Photostat supplied hy M. C. Born.

Figure 15. Karly multitrm-e irtl.ii i,.n srivM;. . i • i-ording

truck used in the San Joaquin Valley, 1933. Courtesy Utiteni Geo-
physical Company.

116

Earthquakes in Kern County, 1952

[Bull. 171

The very first success in using the reflection seismo-
graph to find an oil field in California was that of the
Shell Oil Company (Waterman, 1948). They completed
the first well in the Ten Section field for nearly 1,000
barrels of high gravity oil on June 2, 1936. This com-
pany conducted their reflection seismograph survey of
the area during 1934-35 under the very competent direc-
tion of Dr. W. Hafner. As a result of the reflection
shooting and a careful interpretation of the data, the lo-
cation for the test well was made practically on the apex
of the geologic structure. This was indeed an outstand-
ing accomplishment when one considers that it was done
in a period when the reflection seismograph was still
in its infancy and that the closure of the structure was
only about 200 feet. No surface indication of the struc-
ture exists. This discovery also established that it was
commercially attractive to prospect the deeper zones for
possible oil production in the floor of the San Joaquin
Valley. It also marked the discovery of the productive
Stevens sand zone of upper Miocene age and indicated
that this sand might extend into adjacent areas. Follow-
ing this discovery, all of the subsequent seismic work in
the nearby areas was directed toward the making of con-
tour maps of this geologic horizon.

Only 16 days after the completion of the first Ten
Section well, The Amerada Petroleum Corporation dis-
covered the Rio Vista gas field, which today is Cali-
fornia's largest. This marked the second gas discovery
for Amerada Petroleum Corporation and the sixth suc-
cess for the reflection seismograph in the Great Valley
of California. Shortly after the flow of the Rio Vista gas
field discovery well was established at 81 million cubic
feet of natural gas per day through a |-inch orifice, the
oil operators began a fervid exploration program in the
Sacramento River valley.

The Shell Oil Company's success in Ten Section area
was soon followed by a similar success by the Standard
Oil Company of California when they completed the
first well in the Greeley oil field on December 22, 1936
for more than 2000 barrels of crude oil per day. This dis-
covery also was the direct result of reflection seismo-
graph studies which the Standard Oil Company had
made in the years 1935-36. The Stevens sand zone was
found productive at Greeley thus enhancing the im-
portance of the discovery of this zone at Ten Section
oil field. The general structure of the Greeley oil field
is that of an elongated anticline, about 1 mile in width
and 4 miles in length. It was the first field found on the
northwest-trending Wasco-Rio Bravo-Greeley trend. The
discoveries of Ten Section and Greeley oil fields on the
floor of the San Joaquin River valley stimulated use of
the reflection method in many other parts of the area.
Many of the reflection survej's made prior to Standard 's
discovery of the Greeley structure had been conducted
in reconnaissance fashion. All possible structural re-
versals of these surveys were immediately re-examined
and re-evaluated in the light of these first two oil dis-
coveries. Many completely new detailed surveys were
made, as well as the many older, widely spaced surveys
reshot.

At the time the discovery well at Greeley was being
drilled the Union Oil Company was studying the Rio
Bravo area with the aid of a Western Geophysical reflec-
tion seismograph crew. The historical record indicates

that the Union Oil Company had considered this particu-
lar area favorable as early as 1925 when they drilled
a deep test well on a slight topographic high located
just west of the present Rio Bravo oil field. 'The discov-
ery well was drilled to the equivalent of the Stevens
sand zone established as an oil producer at Ten Section
and Greeley oil pools. The zone was somewhat disap-
pointing to the Union Oil Company in the test well
because it was represented by a hard and dark brown
shale containing a very thin unproductive silty oil sand
lens at a depth of about 8,500 feet. However, the thin
oil saturated sand was interpreted as confirmation that
the test well was on a closed geologic structure as indi-
cated by the reflection seismograph survey. With this
in mind the' Union Oil Company decided to drill ahead
in search of a more favorable oil sand. They completed
the Rio Bravo discovery well, flowing at the rate of
2400 barrells of oil per day, on November 3, 1937 at a
depth of 11,300 feet. The producing zone consisted of
two sands. The upper one was regarded as a new sand
discovery while some geologists thought that the lower
one was the geological equivalent of the Vedder sand,
the main producing zone along the east side of the San
Joaquin Valley. The upper sand was appropriately called
the Rio Bravo sand. All geologists agreed that the sands
were of lower Miocene age. This new well was at that
time the deepest producing well in the world. The suc-
cess of the well increased the demand for reflection seis-
mograph surveys of still deeper geological structures.
The new, deeper studies involved the identification and
study of reflected waves that returned to the ground
surface as late as 2.5 to 3.0 seconds after the blast in the
shot hole. Up to this time reflections for these extreme
depths had not been considered very important and
little or no systematic effort had been made to obtain
them. They are less distinct than earlier ones ; the
periods of the pulses are much longer and the ampli-
tudes are less than those recorcted earlier, resulting in
broader and less sharp pulse recordings. These less well-
defined reflected pulses do not permit extremely accurate
determination of the reflection move out times, which
are essential for the computation of the dip of the reflect-
ing interfaces. However, with more efl'ort being expended
to obtain the deeper reflections, improvements were made
that led to the re-shooting of some areas.

Very soon after the discoverj- of the production from
the Rio Bravo and Vedder sands at Rio Bravo, the
Standard Oil Company of California promptlj^ tested
for these sands at the Greeley oil field. The test well
was completed in the Vedder sand for an established
flow of 14,000 barrels of oil per day marking a con-
siderable improvement over the original production
found in the Stevens sand.

With the discovery of the Rio Bravo oil field and a
second producing zone at the Greeley oil field, a new
structural trend in the floor of the San Joaquin Valley
was found and interested oil operators energetically
focused their attention on it and on the possibility of
finding other such trends. The conditions found at
Greeley and Rio Bravo indicated that this strvictural
trend was plunging northwest.

The next discovery well on the Greeley-Rio Bravo
trend was drilled by the Continental Oil Company. The
well penetrated an oil bearing cherty shale bed of ujiper

Part I]

Geology

117

Miocene age — probably the equivalent of the upper por-
tion of the Stevens sand — at a depth of 9,540 feet. The
well was apparently completed under adverse mechanical
conditions. Initial production was 35 barrels of crude
oil per day. The Continental Oil Company, encouraged
by this experience, promptly drilled another test well
completed in 35 feet of oil sand encountered at a drill
depth of 13,095 feet for a demonstrated flow of 3,385
barrels of oil per day on April 11, 1938. Before com-
pletion, this well was drilled to a total depth of 15,004
feet where it encountered Eocene formations (Erickson,
1948). It established new world records for deep drilling
and deep production.

Subsequent wells drilled in the field indicated that
the "Wasco structure is an elongated dome of very low
relief with approximately 60 feet of closure over a rather
small area. In addition to having the deepest commercial
production, the field also is an outstanding achievement
in the application of the reflection seismograph, as it
was located on a structure of low relief and small area
at very great depth. The 35 feet of producing sand at
"Wa.sco was not the equivalent of either the Rio Bravo
or the Yedder sands at the Rio Bravo oil field, but instead
represented a geological zone slightly higher than the
Rio Bravo sand.

The accomplishment of the Geophysical Department
of the Continental Oil Company in delineating the
Wasco oil field structure has frequently been referred
to as a demonstration that the reflection method was
the most important step in oil and gas exploration. It
has given the oil prospector a means of securing infor-
mation on buried geologic structure almost as definite
as though human eyes had been given the power to look
into the earth. For some time after the Wasco oil field
discovery, no oil pool seemed to be too small or too deep
for the reflection seismograph to find it. We now know
this is not an exact statement because a combination
of conditions may naturally exist which could make
even larger structures if not impossible, improbable to
detect with the reflection seismograph as the only source
of information.

The history of the brilliant successes of the reflection
seismograph in the San Joaquin and Sacramento Val-
leys, as it continued to unfold, is indicated by the accom-
panying list of oil and gas discoveries given in a
chronological order. A complete storj' would require the
interweaving of personal biographies of geologists and
geophysicists and the stories of the seismograph scouts,
who reported to their companies the operations of com-
petitors. One could not exclude a treatment of the scout's
trade secret as to how he knows where to catch the shoot-
ing crew in action. There is that often repeated story
how the party chief had planned secretly to shoot a
particularly important and confidential location on a
Sunday at 4 A.M. only to find the seismograph scout of
a competitor company waiting for them when the crew
reached the spot at daybreak. This amiable fellow though
always an uninvited guest was not completely unwel-
come. He often furnished information about other crews
in the area and frequently would break the startling
news to the unofficial host crew as to where their next
job would be and about how soon they would be moving
to it.

Some important oil and gas fields of the Great Valley of California
found icith the aid of the reflection seismograph.

8.
9.

10.
II.
12.

13.
14.

15.

16.

17.
18.

19.

20.
21.

23.
24.
25.

26.

27.
28.
29.

30.
31.
32.
33.
34.

Buena Vista Lake gaa field Ohio Oil Company

Chowchilla gas field Pure Oil Company

Tracy gas field Amerada Petroleum Cor-
poration .

McDonald Island gas field Standard Oil Company

of California

Ten Section oil field Shell Oil Company

Rio Vista gas field Amerada Petroleum Cor-
poration

Greeley oil field Standard Oil Company

of California _ .

Rio Bravo oil field Union Oil Company

Fairfield Knolls gas fields Standard Oil Company

of California

Canal oil field _ __ Ohio Oil Company

Willows gas field Ohio Oil Company

Wasco oil field Continental Oil Com-
pany . ^

North Coles Levee oil field Richfield Oil Corporation

Strand oil field Tide Water Associated

Oil Company

Paloma oil field Western Gulf Oil Com-

pany,

Vernalis gas field Standard Oil Company

of California

Raisin City oil field Shell Oil Company

Helm oil field _ Amerada Petroleum Cor-
poration

Riverside oil field Amerada Petroleum Cor-
poration

Gill Ranch gas field Texas Company

Lodi gas field Amerada Petroleum Cor-
poration. _,

Thornton gas field Amerada Petroleum Cor-
poration

Ord Ben gas field. Superior Oil Company .__

Moffett gas field Texas Company

McClung oil field Continental Oil Com-

July 1934
November 1934

August I93S

June 1936
June 1936

June 1936

December 1936
November 1937

November 1937
November 1937
January 1938

April 1938
November 1938

Jime 1939

August 1939

January 1941
June 1941

October 1941

December 1941
April 1943

April 1943

July 1943
September 1943
September 1943

pany__ September 1943

Colusa gas field General Petroleum Cor-
poration

Chico gas field Richfield Oil Corporation

Afton gas field Richfield Oil Corporation

Millar (Dixon) gas field Amerada Petroleum Cor-
poration

Corning gas field Superior Oil Company

San Joaquin oil field _ Superior Oil Company

Cuyama oil field Richfield Oil Corporation

Harve.ster gas field Shell Oil Company

Wild Goose gas field Humble Oil 4 Refining

Company

December 1943
January 1944
February 1944

August 1944
September 1944
March 1947
May 1949
February 1950

.August 1951

A study of this impressive list indicates that the rate
of seismic oil and gas discoveries has varied considerably
from time to time in the Great Valley of California. It
is presumed that this rate depends on the ease with
which the undiscovered pools are recognized by seismic
means before drilling. As more and more fields are
found, it becomes harder to find new pools. The time may
come when no simple anticlinal type of structure re-
mains to be discovered in the Great Valley. If discov-
eries are to continue the geologist and the geoph.vsicist
will find it necessary to look for new types of oil traps
such as fault and lithological closures. Success along
these lines may require marked improvements in geo-
physical prospecting techniques and probably some new
ideas about the existing geology. Certainly, continued
success in finding oil and gas in the Great Valley of
California will make necessary close cooperation be-
tween the geologist and the geophysicist. The seismo-
graph may be used to work out suspected extensions oi
known fields, the solutions of faulted structures and
other a.ssociated problems but it seems the days of the
older reconnaissance-type reflection seismograph sur-
veys are rapidly coming to an end. It may be necessary

118 Earthquakes ix Kern CorxTY, 197}2 | Bull. 171

to develop and employ more sensitive methods in order ''<>"• ^y ^'- '^- Hcilan<l. Pniilisher, Pr.-ntiop llaii. In.-.; (2) Oeo-

.- . o 1 ■ -1 1 ■ *i, n t T'^ 11 e iihusiral prospecting for oil, bv L. L. N't'ttloti)ii. I'lililislier, Mc-

to fontl.iue to find new oil pools in the Great Galley ot ;;,-„„..Hili P.,!,,k mmpany. inc.; (H) E.plor.Uon y<-oph„.ics, by

California. .1. J. Jiikn.-iky, puliHsbiMl by Trija PiililisbinK Comiiany; (4) Prac-

ticnl scisiuoioi/i/ and seismic prospevtintj. by L. I). Leet. I*uI)ii.'*hor,

REFERENCES

AiipU'toii Ontiiry-Cruft.s, Inc.; (5) Introduction to geophysicdl

Any more or less complete biblioKrajihy on sei.smic methods prospecting, by JI. B. Dublin. I'ulilisbcr, Mc(!raw-Hill Huok Com-

woul<l be much too long to be given here. Publications on .seismic pany. Inc.; (6) Seismic prospecting for oil. by C. Hewitt l)ix,

prospecting have increased in number rapidly since 1029. A short Publisher, Harper & Brothers; (7) Geophi/sicul case histories,

list of the more recent publications is: (1) Geophysical explora- vol. I, 1948, Society of Exploration Geophysicists.

12. APPLICATION OF SEISMIC METHODS TO PETROLEUM
EXPLORATION IN THE SAN JOAQUIN VALLEY

liY MAURICK ^Kl,.\[t

ABSTRACT

Seismic pidspcctin;,' and eartluiuake seismology are both based
on the huvs of propajiation of seismic waves in the earth. The
first part of this paper ties in the two sciences, and discusses
p.xamplos of earthcpiiikes recorded by seismic parties exploring
for petroleum. Three .seismoKrams, recorileil by a party working
within a few miles of the White Wolf fault, in July, 11)52, clearly
show events which are aftershocks of the major Kern County
earthquake. These events are analyzed and related to aftershocks
recorded by earthcpiake stations.

The second i)art of the jiaper reviews the history and develop-
ment of .seismic exploration in California, from its beginninK, in
the 1920's, to the present day. The refraction and reflection
methods have both been used, but the latter has proved to be the
more effective in discoverinj; petroleum. The earliest reflection
technique employed the "correlation" method. This was replaced
by the "dip" method, and later by "continuous profilinf;".

Those California oil and gas fields whose discovery can be
credited at least in part to the seismograph are indicated. They
are grouped in four periods, which are mainly chronological, but
partly geographical. These are (1) Early gas discoveries (19,'U-
1936); (2) Southern San Joaquin Valley oil discoveries (llKit!-
1944) ; (3) Sacramento Valley gas discoveries (1937-present
time) ; (4) Recent California oil discoveries (1946-present time).

Introduction. Earthquakes in California have been
recorded for at lea.st a century and a half (California
Division of Mines, 1952, p. 1), and it is quite probable
that they have occurred throughout most of geologic
time. It is only during the last century that artificial
earthquakes, produced by the controlled detonation of
explosives under, on, or above the surface of the earth,
have been emploj'cd as a means for exploring fhe sub-
surface structure, especially as applied to the search for
petroleum and other minerals.

Seismic prospecting is therefore derived from earth-
quake seismology, both being based on the same funda-
mental physical laws. Moreover, there are close simi-
larities in instruments and in methods of interpreta-
tion.

Achnowledgments. The writer Avishes to express his
thanks to the Union Oil Company of California for per-
mission to publish this paper, and to John Sloat for
man}' helpful suggestions and comments. The illustra-
tions have been prepared by Robert Bowman, Tajlor
Moore, and Richard Huntley.

Geologists and geophysicists of many oil companies
have contributed much information which has been used
in assembling the lists of seismic discoveries. Apprecia-
tion is expressed for their help.

Thanks are extended to Dr. Beno Gutenberg of the
California Institute of Technology Seismological Lab-
oratory, who has given much information concerning the
Kern County earthquakes of 1952 and interpretation of
earthquakes recorded by seismic parties.

SEISMIC EXPLORATION AND EARTHQUAKES

In both seismic prospecting and earthquake seis-
mology the original data for analysis and subse(iuent
geological interpretation are the seismograms. Seismic
waves originate at the point of detonation of an ex-
plosive charge, or at the origin of movement along a
fault (focus), and are transmitted through or along the

boundaries of formations to the seismometer stations;
by means of proper amplifiers and other electrical and
optical equipment, they activate various traces on the
seismograms.

Although earthquake seismographs are designed to
record most effectively seismic waves produced by earth-
quakes and prospecting seismographs to record explo-
sions in shot holes, each type could, if the disturbance
were close enough or strong enough, record the other
type of wave.

Earthtjuakes have been recorded many times by seis-
mic parties engaged in exploration for petroleum. Sev-
eral parties were working within 100 miles of the White
Wolf fault on July 21, 1952. (BeniofT, Buwalda, Guten-
berg, and Richter, 1952, pp. 4-7). Figure 1, which has
been prepared by the Seismological Laboratory of the
California Institute of Technology, shows the locations
of the White Wolf and other major faults of the South-
ern San Joaquin Valley area. In addition the positions
of the permanent and temporary Institute seismological
stations, and of the epicenters of' recorded aftershocks
are represented.

A Western Geophysical Company of America seismic
party working for the Richfield Oil Corporation in the
southern end of the San Joaquin Valley was within a
few miles of the White Wolf fault during the latter part
of July 1952. Figure 2 is a reproduction of portions
of three seismograms recorded by this party on July 23
and 25 ; their locations are superimposed on figure 1.
Each seismogram shows strong events which undoubt-
edly are produced by aftershocks.

Each illustration is the latter portion of a routine
reflection type seismogram, and on each the earthquake
events were recorded later than a time of about 3.0
seconds, measured from the time of detonation of the
explosion- for the particular seismograms. This is later
than the deepest reflections to be expected from the
shots, and therefore the events must have originated
from other sources.

Dr. Beno Gutenberg, of the Seismological Laboratory,
has examined the seismograms and states that no re-
corded aftershocks occurred at the approximate times
given on figure 2. Because of the wide scattering of the
epicenters of recorded aftershocks (figure 1), it is most
uidikely that any two of the three seismograms would
record earthquakes from the same origin. In fact, seismo-
grams A and B, with the same ground location for the
seismometers, indicate waves with opposite directions of
moveout, and therefore from different sources.

The best determination that can be made of amplitude
of ground motion at the seismometers for the three seis-
mograms of figure 2 is 10"" cm for A and C, and 10"^ cm
for B. These values, which have been supplied by the
Laboratory of Western Geophysical Company of Amer-
ica, from a consideration of seismogram amplitudes,
wave frequency, and sensitivity settings, must be con-
sidered approximations.

Since the distance from the epicenters for these after-
shocks is not known, the magnitudes can only be esti-

(119)

120

Earthquakes in Kern County, 1952

I Bull. 171

Figure 1. Epicenters of Kern County shocks, July 1952 to
March 1053. .4, B, and C, are locations for seismofiraras of fisure
2. Map Inj permission of Seisiiiological Lahoratoiy, Culijornia Insti-
tute of Technology.

mated. They would be 3 to 4 for A and C, 2 to 3 for B.
These are smaller than the magnitudes for figure 1,
which, in general, are 4 or larger.

Present day reflection seismographs are extremely sen-
sitive to low amplitude ground motion. On July 22, when
the frequency of aftershocks reached a maximum, the
ground disturbance caused by them in the area was so
great that it was practically impossible to obtain usable
reflection records, and operations were suspended for the
day by the party.

The seismograms are susceptible to partial analysis
only. If we speculate, in order to carry out the computa-
tions, that the events on seismograms A and C are from
the same origin, the moveout times may be resolved, and
a direction for the wave arriving at the seismometers
determined. It is also assumed that the two spreads of
seismometers are at about the same location. The graph-
ical solution is shown on figure 3.

A resultant time moveout of .060 sec. is indicated, and
the direction of arriving wave is from S 01° E to N 01°
W. Referring again to figure 1, the arrival would either
be from a fault south of the White Wolf, or would
suggest that this fault may have a considerable hade to
the south. The apparent velocity of the wave arriving
at the seismometers is 1200/.060, or 20,000Vsec. This
value lies very close to that of the longitudinal velocity
in granite, 6.34 km. /sec. or 20,800'/sec. ; thus, if the
wave being recorded were longitudinal, the wave path
in granite before entering the superjacent sediments
woidd be about horizontal. No earlier phases of the ar-
rivals on figure 2 are said by Richfield Oil Corporation
to be visible on the earlier portions of the seismograms,
and hence these arrivals are probably longitudinal
waves.

The frequency of the waves is between 25 and 35
cycles per second. This is much higher than that of
earthquake waves which have travelled long distances,

usually I to 4 cycles per second. The absence of the low
fre(iuency components on the seismograms is caused by
the electrical filtering used to transmit reflected energy,
while attenuating frequencies outside of the frequency
range of the reflections. Thus the filter used has a peak
fre(inency between 20 and 30 cycles per .second, and the
much lower frequency components of the earthquake
spectrum have been filtered out. On the other hand the
higher frequencies registered on these seismograms are
not recorded on the usual earthquake seismograms be-
cause they are more rapidly attenuated with long dis-
tances than the lower frequencies, and also because of
seismograph characteristics.

A second example of an earthquake being recorded by
a seismic party is one recorded by The Texas Company,
in Ventura County, California, on June 25, 1947, 12:58
P. M. The earthquake was felt by the observer while
taking a reflection seismogram. and he continued the
recording for several additional seconds in order to re-
cord the earthquake. This seismogram has been described
in a paper entitled An earthquake recorded by reflection
scismoyraph iiititruments by Norman J. Lea (Geophys-
ics, 1948, p. 656) and presented on June 18, 1948, at the
first annual spring meeting of the Pacific Coast Section
of the Society of Exploration Geophysicists, Bakersfield.

The seismogram showed many jihases with difl'erent
apparent velocities. Dr. Gutenberg has examined the
seismogram, and has referred to two shocks felt at Car-
jiinteria within 1 minute, on the same day, the second
having a time of origin of 12:55:54 P. M., and probably
the one recorded by the party. The waves with the small-
est apparent velocity were probably transverse, while
those with the largest were possibly reflections of longi-
tudinal waves from deep layers (bottom of granite?).

One conclnsioii presented in the above paper was that
transverse waves could be recordetl by exploration seis-
mograph equipment. This observation was used as a
possible explanation of lagging reflections observed in
the Salinas Valley.

Because oil-producing areas exist on both sides of the
White Wolf fault, seismic waves, originating from ex-
plosions detonated by geophysical prospecting jiarties,
have been observed after crossing the fault trace. A
seismic line that crossed the trace is shown in figure 4.
This section has a generally northwest-southeast bearing,
and lies between Wheeler Ridge and Comanche Point.
As is typical of much of this area, seismic data are only
fair in quality. Nevertheless, the section shows the abrupt
termination of the continuity of dips, which often is
observed at a fault zone, especially if it comprises a
width of several hundred feet or more. Some of the
steeper dips may be refracted reflections from the fault ;
in this case the plotted dip would not be the dip of the
fault surface.

DEVELOPMENT OF SEISMIC EXPLORATION
IN SAN JOAQUIN VALLEY

General. Although the earliest use of artificially pro-
duced seismic waves dates back into the latter part of
the 10th century, it was not until after 1920 that the
practical development of both the refraction and the
reflection seismograph was realized. The early history of
the development of the methods is described in excellent

Tart I

GnoLooY

121

S 57° Wf A

Figure 2. EarthquaUe events on reflection-type seismograms. By permission of Richfield Oil Corporation.

fashion by Weatherby (1948, 1948a), Salvatori (1948),
and Sc-hriever (1952).

Refraction Prospecting. Refraction was the first of
the seismograph methods to be used to discover an oil
field in the United States; Orchard Dome, Texas Gulf
Coast, was discovered by this method in 1924. The first
refraction work in California was probably done in 1925,
on an experimental basis, and regular survej's were car-
ried out by 1927.

The method had its greatest success in the Gulf Coast
area, where it recorded the large time accelerations at-
tained by seismic paths through high velocity (15,000'/
sec.) salt domes intruded into low velocity sediments
( 5000-10, OOO'/sec). Although the geological section un-
derlying the San Joaquin Vallej- resembles that of the
Gulf Coast in both age and lithology, the method did not

enjoy comparable success, because of the lack of such
surfaces of large velocity contrast within the sedimen-
tary section.

The refraction method was used to best advantage
along the east side of the Valley, where it could trace
the granitic basement surface from the western margin
of the Sierra Nevada under the sediments of the Valley.
Results in the vicinitv of Hanford have been described
(Vaughan, 1943, pp. 67-70).

Refraction shooting in this early period made a valu-
able contribution to our knowledge of the regional geol-
ogy of the San Joaquin Valley. It showed that the syn-
cline of the valley lay not in the center, but close to the
western margin. A minimum depth (Vaughan, 1943) of
30,000 feet is quoted for the depth to basement in the
deep trough immediately east of Coalinga Nose.

122

Eartiiqt^^kes in Kern County, 1052

[Bull. 171

RESULTANT ■ \

I 060 sec N 01° W \,

Surface Corrections,
Seismogrom A = .OOOsec.

Surface Corrections,
Seismogrom C — = .006sec.

Moveout A '■

.032 -000 = .032 sec.

Moveout C '

.066 -.006= .060 sec.

Figure 3. Resolutions of moveouts from seismograms .1 and C,
figure 2, assuming same origin for shocks and same location for
spreads of seismometers.

Reflection Prospecting. The reflection method of seis-
mic prospecting- orio-inated in Oklahoma, where the first
experimental profiles were observed in 1921 (Weather-
by, 1948, 1948a, Schriever, 1952). Bj^ 1927 exploration
for petroleum was in progress, and the first oil field dis-
covered bv this method, South Barlsboro, Oklahoma, was
found in 1929.

The first use of the reflection method in California was
in 1928, but the work was more or less experimental in
nature. Salvatori (1948) mentions that the results were
not very favorable, and that the work was soon dis-
continued.

In 1931, after the method liad led to the discovery of
several oil fields in Oklahoma, anotlier attempt was made
in California. Only a few seismic parties operated until
about 1934. About this time the reflection method led to
the discovery of several gas fields, and, beginning in
1936, of many oil fields.

This proof of tlie value of the method led, of course,
to an increase in the number of parties throughout the
state, and today the number remains high. Figure 5
shows graphically the number of seismic parties em-
ployed ill California from the early 1930 's to the present.
In addition to these parties, there has been at least one
offshore party in operation for most years since 1944.

Technique of Reflection Prospecting. As developed
initially in Oklahoma the reflection method was by cor-
relation. Shot points were spaced at ^ to 1 mile inter-
vals, and neighboring seismograms were correlated, or
matched on reflections with similar appearance, or char-
acter. Depths computed from such correlated reflections
would give tlie configuration of a given geologic horizon.

Tlie method worked quite well in Oklahoma, where
marked differences in lithology between successive for-
mations exist, and continue for long lateral distances.
The method still has an application in certain areas of
the Mid-Continent and Canada, where the formations
are of Mesozoic or Paleozoic age. In California, however,
lithology often varies quite rapidly in short distances,
and reflections do not always maintain their character
for many successive stations.

The correlation method was succeeded by the dip
method. This method was first developed on the Gulf
Coast, where the geologic section resembles that of many
basins of California, and lateral lithologie changes are
also rapid. It was first employed in California in the
early 1930 's.

The dip method emploj's the moveout time of a reflec-
tion across a seismogram. By use of the proper mathe-
matical equations, based on formation velocities and the
geometry of the spread of seismometers and the shot, the

S-E

:• i. :. •.. fJ-W

L«'*'

r« ., -»—

UJ
2
O
M

_jy.»

-J

<
U.

.JUL fa

11.

-J

-^*--..,

1-
I

— ■^•..

Figure 4. Keflectiou seismograph section across ^Vhite Wolf fault.

Tart Tl

Geology

123

dip of the rene('tin<r bed at the point of incidence is
coinptited. As actually used in California, the seismom-
eter spreads were usually spaced alon^ straight lines
(either parallel or normal to the direction of regional
dip), at intervals of a (juarter to half a mile, with the
shot points either in line or offset a few hundred feet.

Dips so computed were projected from one station to
the next, and "phantom" horizons were carried from
station to station, and from line to line. AVhere available,
correlations were used to supplement the dips. The dij)
method, in use driiig most of the 19;i0's and 1940 's, has
led to the discovery of manj' oil and gas fields in the San
Joaquin Valley.

A variation of the line method of laying out stations
is the use of spreads in two different directions (prefer-
ably normal) at a given station. In this ea.se the com-
puted dip components are resolved into a total dip.

YEAH '32 '33 '3* '55 36 '37 '3S '39 '40 '41 '42 '43 '44 '45 '46 '47 '49 '49 '5Q '51 52 '63

Figure 5. Reflection sei.snioKraph parties in California
September 1932 to June 1953.

The relationship between the correlation and dip
method can be readily seen by use of an analogy de-
rived from a consideration of a series of electric logs
from bore holes in an area. Let us suppose that at each
bore hole the ordinary self-potential and resistivity log,
and also a dipmeter survey is available. Logs of holes
may be matched in two ways. If the ordinary logs are
used, they are correlated on character, and the difference
in elevation on a given marker horizon is given by the
difference on the logs (correcting for variation in refer-
ence elevation). The alternative procedure would be to
use the dips from the dipmeter survey at two bore holes,
resolve each into the direction between the two holes,
and project each dip halfway. These two systems are
analogous to the correlation and dip methods of reflec-
tion prospecting.

During the 1940 's, after the floor of the San Joaquin
Valley had been covered in reconnaissance, and the
most prominent structures had been found, the need
arose for a more refined technique to discover the smaller
anticlines, structural noses, faults, overlaps, and uncon-
formities. This method, the one used almost entirely in
California today, is that of continuoKs profiling. This
system, with continuous spacing of seismometer stations
on the ground, and with the stations between each two

successive shot points shot from both points, gives con-
tinuous subsurface coverage. The interval between ad-
jacent shot points is usually between 400 and 1000 feet.

Figure 6 is a reproduction of six successive seimo-
grams, with 4800 feet of continuous subsurface cover-
age, located on the west side of the San Joaqtiin Valley,
between the valley syncline and the Kettleman Hills —
Lost Hills anticlinal trend. Because of the continuity of
coverage, and with the dip all in one direction, the series
of seismograms is, in effect, a subsurface geologic sec-
tion; the vertical scale is not uniform because of the in-
crease of velocity with depth, and the steeper reflections
are, of course, not migrated up-dip. The vertical depths
and the positions of the top of the Pliocene and the top
of the Miocene beds are only approximate.

The seismograms demonstrate the fact that the con-
tinuous profiling method employs both correlations and
dips. Actually both methods are employed in the com-
putations.

The above brief summar.v of the development of re-
flection shooting in California has been concerned with
the evolution of the various types of surface patterns.
This has been accompanied by a continuous imi)rove-
ment in instrument design, both in seismometers and in
recording equipment. The number of stations per spread
has increased from its early value of 4 to 6 to the present
20 to 24, with some 36 and 48 trace units in present day
operation. Multiple seismometers have come into use,
along with the procedure of mixing, both mechanical and
electrical. Amplifiers and filters have been improved to
increase sensitivity and to increase the reflection to
extraneous energy ratio. Improvements in computing
and interpreting methods have kept pace with advances
in field procedures.

RESULTS OF SEISMIC EXPLORATION IN THE
SAN JOAQUIN VALLEY

Introduction. The success of the seismic method as
applied to the search for petroleum sources must be
measured by the volume of reserves and production from
fields whose discovery is credited to the method. The
following paragraphs comprise a brief sunnnarj^ of those
oil and gas fiekls whose discoveries are generally credited
to the reflection seismograph, or towards whose discovery
the method has made a substantial contribution.

Because of the fact that many of the fields here enu-
merated fall in the latter category, many of these fields
are near the line which separates seismic discoveries from
those discovered by other methods, mainly subsurface
geology. Doubtless some readers would not place certain
of these fields here ; a few fields which are not mentioned
here may be considered by some as seismic discoveries.
In general, the fields listecl here are those for which the
company who made the discovery gives at least part
credit to the seismograph. Size of the field, or volume of
reserves or production is not a factor in the selection.
Some of these fields are now abandoned ; others may not
yet have been placed on production, although the poten-
tial has been indicated.

The sources for the data were discussions with mem-
bers of the exploration departments of the various com-
panies concerned, and published records.

124

Earthquakes in Kerx County, 1952

[Bull. 171

Figure 6. Continuous seismograph coverage, west side of San Joaquin Valley. Vertical depths
aud positions of top Pliocene and top Jliocene approximate.

Part 1]

Geology

125

Oil field
^^ Gas field

Figure 7. Location of oil and gas fields discovered by reflection seismograph. Numbers corre-
spond to those used m text. Base map hy permission General Petroleum Corporation.

126

Earthquakes in Kern County, 1952

[Bull. 171

Figure 8. A (above) Pre-discovery seismic map of Ten Section
oil field. Reproduced from "Geophysical Case Histories, rol. I," by
permission of Society of Exploration Geophysirists. li (l)elow) Con-
tours on productive sand, Ten Section oil field. Reproduced from
"Geophysical Case Histories, vol. I," by permission of Society of
Exploration Geophysicists.

Seismic discoveries are here grouped in four divisions,
which are partly chronological and partly geographical,
so as to indicate the trend from the first discoveries to
the present day. All of these fields are credited to the
reflection method, none to refraction.

Early gas discoveries.

Name of field

County

Discovered by

Date

Producing zones

1. Chowehilla

Madera -

Kern...

Kern, Kings,

Tulare
Kern .---

Pure Oil Co

Ohio Oil Co

Trico Oil & Gas
Co.

Standard Oil Co.

Araerada Petro-
leum Corp.

Standard Oil Co.

Amerada Petro-
leum Corp.

November 1934.
July 1934

November 1934.

March 1935

August 1935

April 1936

June 1936

Miocene, Cre-
taceous

(Buena Vista

Uke gas field)

"3 Trico* . .

(upper Plio-
cene*
San Joaquin

4. Semitropic*

San Joaquin..

San Joaquin..
Solano, Sacra-
mento

6. McDonald Island

7. Rio Vista

Paleocene
Eocene

Early Gas Discoveries (1931-36). The first fields in
California to be discovered as a result of exploration by
the reflection seismograph are a series of relatively slial-
low gas fields found between 1934 and 1936. That the
first fields contained reservoirs of gas rather than oil is
due mainly to the fact that in the earlier years of seis-
mic exploration, information much deeper than about
5000 feet was not generally obtained. Moreover, under
the floor of the San Joaquin Valley, the shallow forma-
tions (upper Pliocene in southern part, Eocene-Creta-
ceous in the northern) contain dry gas, rather than oil.

The field generally considered to have been discovered
by the earliest reflection exploration in California is the
Chowehilla gas field, Madera County. (See figure 7 for
location of fields discovered by the seismograph.) As a
result of reflection shooting carried out in 1931 and 1932
(Salvatori, 1948), the field was discovered by the Pure
Oil Company in November, 1934. The seismic work was
of the correlation type. Chowehilla field was not very im-
portant, economically, because of the low heating value
of the gas.

In this and the following sections, an asterisk follow-
ing the name of the field indicates that credit for the
discovery is shared between the seismograph and other

Sou

hern San Joaquin Valley

oil discoveries.

Name of field

County

Discovered by

Date

Producing rones

8 Ten Section

Kern

Shell Oil Co....

Standard Oil Co.

Union Oil Co....

Ohio Oil Co

Standard Oil Co.

Continental Oil
Co.

Ohio Oil Co

Richfield Oil
Corp.

Amerada Petro-
leum Corp.

Tide Water Asso-
ciated Oil Co.

Western Gulf Oil
Co.

ShcUOilCo....

Continental Oil
Co.

Amerada Petro-
leum Corp.

Shell Oil Co

Tide Water .Asso-
ciated Oil Co.

Continental Oil
Co.

General Petro-
leum Corp.

Amerada Petro-
leum Corp.

General Petro-
leum Corp.

June 1936

December 1936.
November 1937
November 1937
January 1938...
April 1938

November 1938
November 1938

April 1939-

June 1939

September 1939.

June 1911

September 1941.

October 1941 ...

December 1941 .

May 1942-

January 1943...

September 1943.

October 1943 ...

July 1944 --

December 1936 .

9 Greeley

Kern

Kern-...

Vedder

11. Canal-

12 Canfield Ranch*

Kern

Stevens

13. Wasco

Kern

Kern...

16. East Coalinga Ex-
tension, Amerada
area*

17 Strand

Fresno.

Kern

Eocene

18. Paloma (Oil)

Kern

Stevens

19. Raisin City

20. Shafter

Fresno _

Kern

Miocene, Eocene
Vedder

21. Helm

Fresno

Kern

22. Riverdale

23 Antelope Hills*

Paleocene.
Cretaceous
Miocene

Miocene

Kern

26. McClung*

26. Burrel - ..

Kern

Fresno

Kern

Miocene

27. Ant Hill*

Miocene

**28. Wilmington

Los Angeles . .

Pliocene,
Miocene

(Doell, 1943, p. 551) Trleo is only partially a .seismic discovery. Mr. Harry Magee
became interested In the area because of gas shows in water wells, and because
of the slight topogaphic expression. The location of the discovery well was based
on seismic work of the St.-indard Oil Company.

** Located in Los Angeles basin.

methods. The number corresponds to its location on
figure 7.

Southern San Joaquin Valley Oil Discoveries (1936-
44). This period was ushered in by the discovery of
Ten Section field, and, as used in this paper, carries to
about 1944. It is overlapped somewhat by the next
period, that of the Sacramento Valley gas discoveries.

Part 11

Geology

127

Sacramento Valley gas discoveries.

Recent California oil discoveries.

Name of field

County

Discovered by

Date

Producing zones

29. Fairfield Knolls

Yolo

Standard Oil Co.

Ohio Oil Co

Standard Oil Co.

November 1937
January 193S . .
January 1941 .-

Eocene

30. Willows

Glenn--

Oetaceoua

31 Vernalis

Cretaceous

Stanislaus

32. Bowerbank

Korn

The Texas Co. . ,

January 1942-.,

San Joaquin

33. Roberts Island...

San Joaquin- -

SUndardOilCo.

August 1942-..-

Eocene

34. Gill Ranch

Madera

The Texas Co.- -

April 1943

Eocene

35. Lodi* -

San Joaquin - .

Amerada Petro-

April 1943

Eocene

leum Corp.

36. Ord Bend

Glenn-

Superior Oil Co..

August 1943---.

Cretaceous

37. Thornton

Sacramento,
San Joaquin

Amerada Petro-
leum Corp.

September 1943.

Eocene

38. Moffat Ranch ...

Madera

The Texas Co. . .

September 1943-

Eocene

39. Colusa .

Colusa

General Petro-
leum Corp..

December 1943 -

Cretaceous

Union Oil Co.

40. Chico

Butte -.-

Richfield Oil
Corp.

January 1944--.

Cretaceous

41. Afton

Glenn

Richfield Oil
Corp.

February 1944.-

Eocene

42. Honker Bay

.Solano

Standard Oil Co.

April 1944

Eocene

43. Millar (Dixon)...

Solano -

Amerada Petro-
leum Corp.

August 1944

Eocene

44. Suisun Bay*

Solano

Standard Oil Co.

September 1944-

Eocene

45. Alpaugh*

Kings

Standard Oil Co.

September 1944-

San Joaquin

46. Corning-

Tehama

Superior Oil Co.

October 1944 ---

PUocene

47. Kirby Hills'

Solano

SheUOilCo.---

January 1945---

Eocene

48. Maine Prairie

Solano

Amerada Petro-
leum Corp.

March 1945

Eocene

49. Cache Slough* --

Solano.

Standard Oil Co.

March 1945

Eocene

50. Dunnigan Hills* .

Yolo

The Texas Co.- -

February 1946- -

Cretaceous

51. Winters

SheUOilCo.--..

February 1946.-

Cretaceous

52. Durham -

Butte

Standard Oil Co.

July 1946

Eocene

53. Pleasant Creek*

Yolo

SheUOilCo

December 1948 .
February 1950- -

Cretaceous

54. Harvester*

Kings

San Joaquin

55. Wild Goose

Butte

Honolulu Oil
Corp.. Humble
Oil4Rfg.Co.

September 1951-

Cretaceous

56. Freeport'

Sacramento- --

Sundard Oil Co.

May 1952

Cretaceous

57. Sutter

Sutter

Richfield Oil
Corp.

August 1952

Paleocene

58. Beehive Bend.- --

Glenn

Sunray OU Corp.

May 1953.

Cretaceous

The correlation method of seismic prospecting had by
1936 been replaced by the dip method, since the correla-
tion method was not successful in most areas. It has been
the dip method which lias led to the discovery of most
of the fields described in this paper.

The first discovery of this period, and the first dis-
covei'y of an oil field in California by the reflection seis-
mograph, was of Ten Section field, Kern County, on
June 2, 1936, by the Shell Oil Company (Waterman,
1948, pp. 551-553). Figures 8A and SB are reproduced
from Mr. Waterman's paper, and show a pre-discovery
seismic map for comparison with contours on top of the
productive sand.

The discovery of Ten Section was followed within the
next few years by that of many other oil fields, prin-
cipally in Kern County.

The Stevens and Rio Bravo — Vedder sand zones of
Miocene age, have been the sources of much of the oil
produced from beneath the San Joaquin Valley. The
Stevens sand was first produced in the Ten Section field.
The Rio Bravo — Vedder sand was first produced under
the valley floor from Rio Bravo field (Sloat, 1948, p.

Name of field

County

Discovered by

Date

Producing zones

59. l^effingweU*

60. San Joaquin*

61. West Montalvo" .

62. South Cuyama- . .

63. Calder Corners*

Los Angeles . -

Fresno

Ventura

Santa Barbara

Kern

Standard Oil Co.
Superior on Co..
Standard Oil Co.

Richfield Oil
Corp.

General Petro-
leum Corp.

The Texas Co...

Superior Oil Co..

Union Oil Co.---
Union Oil Co.-- -
Tide Water Asso-
ciated Oil Co.

February 1946. .

March 1947

April 1947

May 1949

May 1949 -

January 1952...
August 1952

September 1952-

March 1953

April 1953

Pliocene
Eocene

Pliocene, Mio-
cene
Miocene

64. Goose Slough*

Kern

ment area)*

66. Jesus Maria*

67. Tejon Flats*

Santa Barbara

Kern

Kern - --

Miocene
Miocene

worth area

569) ; this production, from below 11,000 feet, was at
that time the deepest in the world.

Wilmington field, Los Angeles County, is included in
the accompanying chart becavise of its early time of dis-
covery. The discovery of this field, by far the largest
found by the seismograph in California, with an esti-
mated ultimate production valued around $2,000,000,-
000, resulted from drilling a structure outlined by re-
cordings from only four stations, and supplemented
before drilling by 16 others (Salvatori, 1948).

Sacramento Valley Gas Discoveries (1937 -June 1953).
Gas had been discovered by the seismograph in the San
Joaquin and southern Sacramento Valleys previous to
this period. Discoveries in this period resulted in many
gas fields, almost half of the total number of fields con-
sidered in this paper. Most of these are located in the
Sacramento Valley, with a few in the San Joaquin.
Whereas most of the early discoveries are credited al-
most entirely- to the reflection seismograph, many of the
more recent are shared between it and subsurface or
stratigraphic methods.

The names assigned to the more recent of these dis-
coveries are geographical descriptions, not the official
field designations, which have not yet been assigned.

Recent Period of California Oil Discoveries (1946-
June, 1953). This period is characterized by a lower
frequency in the discovery rate, as discoveries become
smaller in size and harder to find. On the other hand,
two of these fields. West Montalvo and South Cuyama,
are comparable in size to most fields discovered during
the period of southern San Joaquin Valley exploration
(1936-44).

The continuation of seismic discoveries in the San Joa-
quin Valley and in the other oil and gas provinces of
California is definitely indicated by the fact that dis-
coveries are being made practically up to the date of
this paper, and by the maintenance of a high level of
seismic exploration.

PART II— SEISMOLOGY

INTRODUCTION

PART II comprisps the seisniolojiic record of the Keru
County earthquakes, computations using the voluminous
data recorded by seismographs, and the conclusions of
seismologists regarding the origin of the earthquakes,
the mechanism of faulting, and relationships of the
White Wolf fault to the regional fault pattern. This
Part is the work of the Seismological Laboratory of the
California Institute of Technology, with the exeeption
of introductory statements on the history of earthquakes
in California by Dr. VanderHoof (Part 1 1-2) and a con-
cluding paper on records obtained from strong-motion
seismographs by the United States Coast and Geodetic
Survey (Part 11-12). Preliminary to a complete tech-
nical discussion of re.sults obtained from instruments
at the Seismological Laboratory (Part II-6, 7, 8, 9) are
introductory chapters on principles of the science of
seismology and definitious of its terms (Part II-l), his-
tory of earthquakes in the Valle.v (Part II-3), develop-
ment of instruments used in recording earth((uake waves
(Part II-4), and the location of seismograph stations in
California (Part II-5).

Probably no earthquake in history has had as com-
plete seismological coverage as the Ar\in-Tehachapi
shock on Juh- 21, 1952 and the hundreds of succeeding
aftershocks. This was largely because of the Seismologi-
cal Laboratory at Pasadena, which not only obtained
seismograph records at the Pasadena station and semi-
permanent stations in the southern Sierra Nevada, but
also, within a few hours after the major earthquake,
set up a series of portable seismographs in the earth-
quake area as basis for a special recording program on
a scale never before undertaken. The instrumental
records provided data for computation of magnitudes,
locations of epicenters and foci "hitherto not available
for any earthquake." The approximate distribution of

aftershock foci around a seismic source has been deter-
mined for the first time. The Arvin-Tehachapi is the
"first major eartluiuake for which a sufficient number
of highly sensitive seismogra])hs with time good to the
nearest tenth of a second were in operation at short
distances in different directions." This permitted locat-
ing each shock within a few miles — in depth and in
epicentral position — and times within a fraction of a
second. The extraordinarily detailed local coverage was
supplemented by data obtained from many distant seis-
mograph stations in all parts of the world.

Drs. Gutenberg, Benioff, and Richter, as the result
of evaluation of the data to mid-1953, have been able to
develop new and revised concepts concerning such
things as foreshocks in major earthquakes, characteris-
tics of the release of strain in aftershocks, the radiation
of energy in a shock, the relation of first motion of the
ground to direction of slip on the causative fault, and
the speed and direction of progression of faulting. They
have arrived at quantitative values for the strike and
dip of the plane of the White Wolf fault, the rate of
progression of faulting, and direction of slip in the
fault plane, from the instrumental data. These measure-
ments and concepts have made it possible to derive
additional conclusions as to the mechanism and strain
characteristics of the White Wolf fault (Part 11-10)
and the relationship between activity on that fault and
the pattern of faulting and other geologic structure's in
the region (Part 11-11).

Part 11-12 summarizes the conclusions of the irnited
States Coast and Geodetic Survey based on records ob-
tained from their strong-motion seismograph and in-
cludes a map showing intensities felt in California, ac-
cording to the Modified Mercalli Intensitj- Scale.

(129)

CONTENTS

Page

1. General introduction to seismolouy, by II. Benioff and B. Gutenberg 131

2. The major earthquakes of California: a historical summary, by V. L. Vanderlloof 137

3. Seismic history in the San Joaquin Valley, by C. F. Richter 143

4. Seismograph development in California, by II. Benioff 147

5. Seismograph stations in California, by B. Gutenberg 153

6. Epicenter and origin time of the main shoelc on Jul.y 21 and travel times of major phases, by B. Gutenberg

7. The first motion in longitudinal and transverse waves of the main shock and the direction of slip, by B.

Gutenberg 165

8. Magnitude determination for larger Kern County shocks, 1952; effects of station azimuth and calculation

methods, by B. Gutenberg 171

f). Foresliocks and aftershocks, by C. F. Kichter 177

10. Mechanism and strain characteristics of the Wliite Wolf fault as indicated bv the aftersliock sequence, bv

H. Benioff 1 ' 199

11. Relation of the White Wolf fault to the regional tectonic pattern, by H. Benioff 203

12. Strong-motion records of tlie Kern County earthquakes, by Frank Neumann and William K. Cloud 205

(130)

1. GENERAL INTRODUCTION TO SEISMOLOGY

Hy }l. Hi-:nh)I'1'^ and B. Gutenbkrq

ABSTRACT

This paiior prospiits sfatoments of the prosent status of .scmip of
the inoro iini'ortaiit cimcepts in seismology. It ineludes discussions
of the origin anil mechanism of earthquakes, energy sources, dis-
tribution of earthquakt's in space and time, eiiaracteristics of seis-
mic waves, carthiiuake magnitude and intensity, nature of faulting,
methods of locating foci, aftershocks, wave paths and tsunamis.

Earthquakes result from stresses which accumulate
within the outer 400-niile shell of the earth. The origins
of these stresses are still obscure both as to the source of
energy and as to the mechanism by which this energy is
converted to strain energy. The energy source is gen-
erally assumed to be of thermal origin (radioactivity,
cooling, etc.) although other sources such as gravita-
tional forces may also be active. The mechanisms which
have been suggested for transferring thermal energy into
mechanical energy or elastic strain are convection cur-
rents, change of phase or state, diffusion processes, ex-
pansion, or contraction. Gravitational return to equilib-
rium from disturbances produced in the past may also
play a role. However, these mechanisms appear to the
authors to be generally inadequate to explain all the
known observations, so that other as yet unknown sources
may also be active.

The large majority of earthquakes have a tectonic
origin although, in the vicinity of volcanoes, shocks are
produced by the processes of volcanism. These are classed
as volcanic earthquakes and, in general, are small and
shallow. Tectonic earthquakes are generallj^ assumed to
be generated by release of strain in crustal rocks,
brought about bj- sudden slippage on faults iia accord-
ance with the elastic rebound theory of Reid (1910).
Other mechanisms may also be involved. The number of
earthquakes which have been definitely related to ob-
served fault displacements is very small. According to
C. P. Richter there are only about 20 instances through-
out the world in which faulting in association with an
earthquake has been adequately established by field ob-
servations. Four of these occurred in California — on the
San Andreas fault in 1857 and 1906, in Owens Valley in
1872, and in the Imperial Valley in 1940. There are

SHALLOW EARTHQUAKES MAGNITUDES > 80 SINCE 1904.

Figure 1. Distribution of great shallow earthquakes.

O FOCAL DEPTH. h=70-300KM. MAGNITUDES > 7.5
• FOCAL DEPTH. h=30O-65OKM MAGNITUDES > 7.0

LARGE INTERMEDIATE AND DEEP FOCUS EARTHQUAKES SINCE 1904

Figure 2. Distribution of deep-focus earthquakes.

many other instances, such as the Arvin-Tehachapi
(Kern County) 1952 earthquake, in which the available
evidence for faulting is not entirely convincing; and
still others about which there is no information. Many,
especially small earthquakes, such as the Long Beach
19:^3 and the Santa Barbara 1925, produced no visible
surface evidence of faulting. Others occur beneath the
ocean and in consequence the surface traces are not ob-
servable, if they exist.

The largest number of eartlKjuakes and those with the
greatest energy occur in the upper 40 kilometers of the
earth's crust. Deeper earthquakes occur with decreasing
frequency down to the 250 km level ; below that level
the frequency of occurrence per unit depth interval be-
comes approximately constant to a depth of 700 km.

Although earthquakes may occur anywhere on the
earth, the great majority are concentrated in the cireum-
Pacific belt (fig. 1), which includes at)out 80 percent of
the shallow shocks, 90 percent of shocks occurring at
depths between 60 and 300 km, and nearly all of the
deeper ones (fig. 2). Nearly all of the remaining large
intermediate and shallow shocks occur in the ]\Iediter-
ranean and trans-Asiatic belt. In addition, only one deep
earthquake (in southeastern Spain on lyiarch 29, 1954,
depth 640 km, magnitude 7.1 ±) has ever been observed
in this belt outside the area sTirroundiiig the Pacific
Ocean. A smaller number of earthquakes including a few
major shocks occurs along the principal ridges of the
Atlantic, Arctic, and Indian Oceans. The Pacific basin
and the continental shields are very nearly inactive. For
further details see Gutenberg and Richter (1954).

Accurate information regarding the distribution of
earthquakes in time and space has been available only
for the last 50 j'ears, since instrumental observations
have been possible. Although this is a very short interval
of geologic time the observations show that, in general,
activity does not proceed at a uniform rate. Thus since
1904 earthquakes of magnitude 8 and larger have oc-
curred in about five active periods (fig. 3) of decreasing

(131)

132

Earthqt'akes IX Kerx Cot'xty, 1!)52

[Bull. 171

fi-i

»n 1 1

(

55l

d <

>'^

^ 50. I.._ J

^■'i

o ^

.i/-'

^ i

~: ■ :J4f:"

fo (

)

■yj' •

.■

^ ■

^ ?S

in!

-■ 1 .r

AG 8 6

AG 8 3
4AG 8 0

■i /

■'

n! d

1900 1910 1920 1930 YEAR 1940 1950 I960

tLASTIC STRAIN-REBOUND CHARACTERISTIC SEQUENCE OF WORLO SHALLOW EARTHQJAKtS

MAC i «0. SINCE 1904

FnUHE 3. Strain release as a function of time for all great
eartlii|iiakes since 1".)04.

amplitude and duration (Benioff, 1951). IMoreover, there
are different patterns of world activity for each of the
three major depth classifications (Benioff, 1951).

Any disturbance at a jjoint within a solid produces
two principal wave types — longcitudinal and tranversc —
(figs. 4, 5) which proceed with different speeds depend-
ing upon the physical properties of the rock. Longitudi-
nal waves (P, push-pull) always travel faster than
transverse waves (S, shear-displacement at right angles
to direction of propagation) (fig. 6). The energy is not
propagated uniformly in all directions and the direc-
tional pattern of longitudinal waves is not the same as
that of transverse waves. The direction of maximum
radiation for transverse waves is approximately at right
angles to the plane of the fault, whereas the longitudinal
radiation pattern has minima in the direction of the
fault plane and at right angles to it. In earthquakes the
amplitudes and periods of the transverse waves are
usually greater than the amplitudes and periods of the

... 1

DEPTH, KM 1000

2000

3000

4000 5000 6000

1 1

-n

VELOCITY

LONGITUDINAL WAVES ( P, K )

TRANSVERSE WAVES (S)

-6-/

^y^t

km/sec

—

— 5/ /ll —

—

7 —

—

1 /lO-

^lUP)

—

1/ g

- —

—

ANT

. 1 _

" ^ " "-

~ 1

longitudinal waves. Those P and S waves which arrive
at the surface near the source produce two additional
waA'e types known as Rayleigh waves and Love waves.
These travel along the surface only. The energy released
in the greatest earthquakes is very roughly equivalent to
10,000 of the original atomic bombs; the smallest earth-
quakes recorded near the source liberate approximately
10"" times as much energy as the largest shocks. With
the present available data it is not possible to calculate
accurately the energy of earthquakes and conseiinently a
magnitude scale has been devised by Professor Kichter
(1935) which is based on the maximum recorded ampli-

1-kr

— %m^' ' ;;.\1^;waW.W^^

Figure .5. Seismogram of aftershocU, August 11, 1952, 13:22:12,
recorded hy Benioff vertical capacity .seisnioKrajili at Pasadena ; dis-
tance 12ri± l;ni. Time marks are 1 minute apart, successive lines
are 15 minutes apart.

P ^ PP PcP

FiGUUE 4. Wave velocity as finidioii of dcplli in llie earth.

Figure 6. Record of main shock July 21. 1052, at Palisades,
New York, distance 35.7°. Columbia I'niversity vertical seismo-
fjraph. To = 10 sec, Tj; = 75 sec. Time marks every minute, suc-
cessive lines are 1 hour apart. Motion down on record corresponds
to motion up of the ^''ound (compressions).

tude of a standard torsion seismograph located at a dis-
tance of 100 km from the source for shallow earth-
(|uakes. Tables have been constructed for calculation of
the magnitude at all epicentral distances and for various
focal depths and for several types of waves. An approxi-
mate relationship between the magnitude M and the
energy E liberated as seismic waves has been given
by Gutenberg and Kichter (1942) in the form
log E = .4 -f BM.

The constants ^1 and B have been revised a number
of times. The values for the constants (A r= 12, 7> ^
1.8) given originally by Gutenberg and Eichter (1942)

Part TT]

Seismology

IS.*?

appear to lead to values of the eiierixy whieli are exces-
sive. For slioeks of iiia^'iiitude <;reater than perhaps 61.
A = 7.5, B = 2.0 are preferable. The constant B is not
necessarily 2. as miiiht be cxiieeted. because the period
and duration of tlio wave trains both var,y with the niaii-
nitude. Ai'tually a better approximation to the relation
between inajiiiitude and enerfry wonkl nnd(nibtedly re-
quire terms of hijiher order, or ditTerent values of .1 and
B for small and for large shocks.

Seismic intensity refers to the violence of shakinfr
at any piven point. On the other band, the scales of
intensity now in use, such as tlie modified ]\Iercalli
scale, are based on th.e eft'ects or destructiveness produced
by the <rround vibration at particular points on avail-
able structures and thus are only ronphly related to
the actual intensity, ilagnitude refers to the eartluiuake
as a whole and is a constant for each shock, whereas
Jlercalli intensity varies from point to point dependinfr
on the distance from the source, the nature of the struc-
tures involved, and the density of population. The de-
structiveness of an earthquake depends upon the energy
released.xthe focal depth, the distance from the source,
the relative orientation of source and structure, the
nature of the ground, the spectrum (distribution of
energy with respect to period) of the source, the spec-
trum (distribution of vibration period) of the struc-
tures, as well as the type of structure.

Although the relationship between the magnitude ]M,
the modified ^lercalli intensity lo near the epicenter,
the maximum acceleration «„ and the radii R of iso-
seismals (assumed to be circular) are too complicated to
be expressed exactly, it is possible to construct a rough
tabulation of the way in which these quantities vary
with the magnitude. Such an approximation is given in
table 1 which is based on calculations by Gutenberg and
Richter (1942) for average conditions in southern Cali-
fornia. Actually, the isoseismals are quite irregular in
■ shape, depending upon the ground conditions and under-
lying structures as well as the strike, dip, extent, and
depth of faulting. Similarly, the other tabulated quanti-
ties are subject to large variations from the assumed
averages.

Tahle 1. MoiliiJed ilercnJH intensity /„ near the epicenter, maxi-
iniini nrrelerntinn Ho in cm/sec' and its ratio to gravity g, and radii
r in km for isoseismals corresponding to various intensities 1, as
function of magnitude .1/. After results of Gutenherg-Kichter (19-i~J
for average southern California earthquakes.

STATION

2.2

1.5

0.001

2.8

0.003

4.5

0.01

6.2

0.04

7.8

130

0.13

9.5
460
0.5

11.2

1670

1.7

cm/sec'

rforI=lH

r for 1=3

rfor 1=5

rfor 1=6

rfor 1=7

rfor 1=8..

r for 1=9

50
30

110

200

120

400
220
100
60
40
25
10

750

400

200

130

km
km
km
km
km
km
km

TRANSVERSE IS)
.t. .t. .(. .t.

The horizontal extent of faulting in a given earth-
quake varies from perhaps a few feet in the smallest
earthquakes to at least 400 km, as observed in the San
Francisco 1906 shock. The extent of faulting downward
has never been determined but probably occasionally

TYPICAL WAVE PATHS. SHOWING DISTANCES TRAVELED IN 13 MIN.

FlorRE 7. Soino t\]^io;il w;i\o jiaths. T., = .mirfnro waves. '\V2
= surface waves over the greater arc. Dotted lines are continuation
of the wave paths after i;'. niiuutes following the origin time of the
earthquake.

amounts to more than 50 km since aftershocks have been
observed to occur with a range of depth of this magni-
tude. The largest observed fault displacements occurred
during the Yakntat Bay earthquakes in Alaska in 1899
when the vertical relative slip reached a maximum of
15 meters (47 feet).

Although some faults have very nearly vertical slip
surfaces, nearly all orientations are observed. "Wherever
a fault intersects the surface of the earth the intersec-
tion produces a feature known as a fault trace.

Instrumental evidence indicates that faulting is initi-
ated at depth at a point called the focus or hypocenter,
from which it proceeds along the fault surface in two
dimensions, presumably with ditTerent speeds. Faulting
in the direction of slip must proceed with a s])eed less
than that of the compressional wave and faulting per-
pendicular to the direction of slip must proceed with a
speed less than that of the shear wave. The point verti-
cally above the focus or hypocenter is designated the epi-
center. It is usually not on the fault trace except when
the fault siu'faces are vertical. The point of origin calcu-
lated from seismograms always refers to the hypocenter
and usually does not correspond with the region of
maximum energy release. The direction of slip may be
horizontal, vertical, or a combination of the two. Under
favorable conditions the first motion of the ground at
an observing station is directly related to the direction
of slip.

Slipping of the fault produces the two principal types
of body waves, longitudinal (P) and shear (S), as
well as Rayleigh waves and surface shear (Love) waves
mentioned earlier. The latter waves travel along the
surface with amplitudes decreasing rapidly with depth
and with speeds equal to or less than the S-wave speed.
In most cases, destructiveness produced by shear waves
is greater than that produced by other types of waves.
Since all waves travel from the origin over minimum
time paths and since for most of the earth the velocity

134

Earthquakes in Kern County, 1952

[Bull. 171

increases with depth, most of the paths within the earth
are curved with the concave side up (fig. 7).

The precise determination of the location of the source
involves trian<iulation based principally upon the time
of arrival of the first waves at a number of observinjx
stations. A rough determination at a single station can
be made using the direction of the first motion of ground
as exhibited by a three component seismograph assem-
bly together with time interval between the arrival of
the compressional (P) and shear (S) waves and other
phases with established travel times. For exact deter-
mination of position and origin time, eifects of local
.structure between source and station must be taken into
account. With a dense distribution within 200 km of
the epicenter of stations furnishing accurate data, the
time of origin can be determined within ±0.2 second
and the position of the epicenter can be located within
a radius of ±3 km provided that there are no serious
systematic errors. The depth is not so accurately deter-
mined. In shallow sliocks the error may be quite large.
Thus, for shocks at a depth of Ki km the error may well
be ±6 km in favorable cases. At a depth of 50 km it
may be as high as ±20 km.

According to the elastic rebound theory, an earth-
quake is initiated at a point where the gradually accu-
mulating stress becomes equal to the strength of the
rock and so produces a slip (fig. 8). The final increment
of stress may be a result of some external force such as
a tidal stress, a change in tlie barometric pressure,
loading by precipitation, etc. Frequently large earth-
quakes are preceded at intervals of hours or days by
foreshocks. Usually the foreshocks are small. They may
be single or multiple. There is no way of knowing which
of the many small shocks occurring in a region is a
foreshock until after the occurrence of the principal
shock. Clearly the occurrence of a foreshock increases
the stress on the fault in its neighborhood and thus has-
tens the break of the main shock. Fairly frequently a
large earthquake is followed within a few liours or days

a D_

_a b

I ^ ' ■

> " ' '

a b

ABC

Figure 8. Schematic illu.stration of rebound theory. A, un-
straineil blocks; /{, strain condition preceding earthqualie ; C, con-
figuration just after earthqualie.

or months by another of similar or greater magnitude
in the same region. Every large shallow earthquake is
followed b.v a great number (many thousands) of after-
shocks (fig. 9). Occasionally swarms of small earth-
quakes occur without any principal large shock. In deep
earthquakes the number of aftershocks appears to be
mucli smaller than in shallow shocks although there may
be a large number of small aftershocks which on account
of their greater distance from sensitive instruments
escape detection. The lack of surface waves in these
earthquakes also makes their detection more difficult on
seisniograms written with older, long-period seismo-
graphs. The frequency of aftershocks is usually highest
immediately following the principal shock and it falls off
rapidly (fig. 10) with time so that usually the sequence
ends within 1 or 2 years. Studies of aftershock sequences
indicates that they are produced by elastic afterworking
of the fault rock. With sensitive instruments located
not too far from the epicenter, the ground is observed to
be in continuous motion for intervals of one or more

^TWt

11 I "

r<«w»>it«MMM^»« .Naa«MWf>i>iM>n« ■■|||H|||M|UUn| [UUIII

' :■ l*W*i»i''i;1l.'iM'i»IINMtRlt|lil»iiitti»«'*«,,iW«»ii ',i,i..i, WHi|k(,*..'*'«llW*...",

',U"I!I1

PALOMAR SHORT PERIOD VERTICAL
1952 July 21, I 1:52:51.8

Fun RK 9. I'aloniar high magnification seismogram covering first 4 hours after main
shock. Successive lines are 1.5 minutes apart. Time marks are every 30 seconds.

I'art II'

Seismology

135

.1 I, I

V. \\^ *^'*Ma,-' ■^.- -,VA\"*V ^^"^. Vv^

y^^jgji

±A:

-t++4

I. ' III ,

3r=:

ffl

Miw^

Figure 10. Portion of a spismoRram written by NS torsion seismograph at Santa
Barl)ara. July 23-24, l!tri2. Time marks are 30 seconds apart, successive lines are 15
minutes apart. LarKe i)hase in each shock is S. Surface waves are not yet well devel-
oped in these epicentral distances of about 100 to l."iO km. Magnitude of largest .shock
is about 4J.

days following a great earthquake as a result of the
high frequency of aftershocks.

In great eartluiuakes the faulting movement may last
for 1 minute or even more. The duration decreases
rapidly with decreasing magnitude. With increasing
distance from the source, the wave movement increases
in duration so that at a distant station an earthquake
which was generated by, say, 1 minute of faulting may
record for 8 or more hours. Waves traveling within the
earth and reflected from the various boundaries (fig.
7) including the surface of the earth have been ob-
served to travel for more than one hour. Surface waves
have been observed after seven complete circuits around
the earth more than 12 hours after they started.

Observers in the vicinity of the epicenter of great
earthquakes occasionally see large waves in the ground.
However, these waves never leave any visible evidence
of their movement such as failures in concrete, etc. and
we are of the opinion they are an optical effect pro-
duced by the large fluctuations in atmospheric density
with consequent variations in the index of refraction as
a result of large vertical vibrations of the ground, par-
ticularly those connected with slow Rayleigh waves.

The f requencj' range of earthqtiake waves extends from
some high frequency limit in the audible range to waves
of at least 8 minutes and, perhaps, even nearly 1 hour
period. The audio frequency components are often heard
by many persons in the vicinity of the epicenter and can
be ver}' loud. These components however are rapidly

damped or scattered with distance from the source.
Sound waves travel within the earth with speeds up to
20 times that of sound waves in the air.

Large earthquakes in the vicinity of coasts sometimes
produce great ocean waves known as tsunamis. Usually
the tsunamis occur only in regions adjacent to the great
oceanic trenches. Several causes for the origin of tsuna-
mis have been suggested. Since tsunamis have appeared
with earth((uakcs whose epicenters lie up to 100 miles
inland it has been assumed that these ocean waves were
initiated by land slides on the steep slopes of the neigh-
boring trench. Thus the source of the tsunami, on this
hypothesis, is displaced both in space and time from
that of the earthquake. In places where the fault slip
extends to the ocean bottom in the region of a trench a
dip-slip movement should produce a tsunami. If a trench
lies within the epicentral region of a great earthquake
it is possible that tsunamis may be generated by the
long period surface waves. In the open ocean the dis-
tance from crest to crest of the tsunami waves may be
several hundred kilometers. They travel at speeds rang-
ing from about 70 meters j)er second in 500 meter deep
water to 300 meters per second in 9 km deep water.
Their periods usually range from about 15 to 30 min-
utes. Their amplitudes in the open ocean are small so
that tsunamis are frequently mi-ssed by ships; upon
approaching shore the amplitudes may rise to 20 meters
or more.

2. THE MAJOR EARTHQUAKES OF CALIFORNIA: A HISTORICAL SUMMARY

By V. L. VanderHoof

That portion of the eartli's enist that lias come to be
known as California has been experiencing earthquakes
since the beginnin<? of time. At some periods in the
greolopric past, earthquakes were less freciuent than at
others. It may be that we are now living;- at a time of
greater frequency, for it is likely that we are still wit-
nes.sing mountain building: of the episode which geol-
ogists call the Coast Range Revolution.

California (and the whole Pacific coast) is part of one
of the great mobile belts of the earth, and relative turbu-
lence and instability of the crustal i-ocks are character-
istics only too plain during a major shock. It nnist be
understood that this turbulence and instability cause
changes in the landscape scarcely noticeable during the
life span of a human being — probably less than effects
produced by other concurrent geologic agents, such as
erosion, volcanism and atmospheric circulation.

But the enormous forces (vastly greater than man-
made nuclear explosions) released by a major earth-
quake do manifest themselves suddenly and the changes
they often produce on the earth's surface are immedi-
ately apparent. This catastrophic aspect is seen in a few
other geologic phenomena, notably volcanoes and land-
slides, and all liave terrified man throughout the ages.

Much as science considers and evaluates earthquakes
as normal natural phenomena capable of being recorded
instrumentally with great precision, it is vitally neces-
sary to consider them as another burden, when destruc-
tive, that man has to bear during his brief tenancy of
the planet Earth. Literate people no longer impute
supernatural causes to quakes, nor do they regard them
as some sort of penance imposed for group sin. Rather
they are looked upon as recurrent hazards, like fire and
hurricanes, to be considered when a house is to be built.
For after all, it is man's house that is shaken down. Lest
we forget, he is the only earth inhabitant that constructs
devices that unemotionally record the character of the
quakes, and it is he that writes about them.

Scales of intensity reflect man's interest in what hap-
pens to him or his works: "destructive, generally felt,
fall of chimneys," and the like are phrases useful in
assessing the size of an earthquake where no instruments
are located. But if we view quakes as a natural phenom-
enon, they are not destructive in the absolute sense ;
they merely cause a rearrangement, of greater or lesser
degree, of certain components of the earth's crust,
whether bricks, bric-a-brac, soil or bedrock.

In writing the history of an earthquake, as is done in
the present volume, seismologists rely primarily on seis-
mographic data to evaluate the fundamental nature of
the shock, but it is very necessary to have also the testi-
mony of eye (or sense) witnesses who are at or near
areas of greatest intensity. For it is they who can de-
scribe the effects on the works of man, something the
instruments cannot do. But in comparing numerous ac-
counts of witnesses to the same event, we at once notice
that there is disagreement over both major and minor
details, as one would expect. This "phenomenon of un-
certainty" is augmented in witnesses of an earthquake

because the witnesses' sense organs are affected at the
same time as his environment. People, like structures,
are affected differently, but nonetheless, the net impres-
sion is relative, not absolute.

;Man's works are likewise affected according to their
orientation and hence a witness is bound to be influenced
by the behavior, during a quake, of any building he is
in or near. One cannot help conjecturing that a person
standing on a featureless plain could give a better ac-
count of an earthquake than one in a city! Professor
Branner of Stanford University once looked" into a phase
of this matter and wrote an account Avhich he entitled
"The untrustworthiness of personal impressions of di-
rection of vibrations in earthquakes." * Branner pointed
out that "my own observations (of 23 years) of things
overthrown lead me to attach very little or no impor-
tance to the direction in which they fall. In the Califor-
nia Earthquake of 1906, a vast amount of data was col-
lected on this subject. Statues, monuments in cemeteries,
chimneys and loose and unstable objects generally, were
thrown in every conceivable direction. The direction in
which such things fell was determined much more fre-
quently by some accident of mounting, such as the shape
of its base, than by the direction of any particular earth-
quake waves." And Branner concludes liis account with
this: "In.strumental records show that the directions are
many and the movements complex. Out of such entangled
movements it seems quite impossible for our uncertain
impressions to gather trustworthy conclusions regarding
the location of epicenters."

In the Hereford, England, earthquake of 1896, four
hundred and sixty-nine observers made notes of their
impressions of the direction of vibration. Dr. Charles
Davison, who described this quake, noted that "when
those directions are plotted on a map of the district, it
is seen at once that they are either nearly parallel or
perpendicular to the roads in which the observers were
living; that is, the ajiparent direction of the shock was
at right angles to one of the principal walls of the house.
This, of course, is a result to be anticipated, for, what-
ever the direction of the earthquake motion, a house
tends to oscillate in a plane perpendicular to one or the
other of its walls."

The present writer happened to be at Saunders Ranch
in Tejon Canyon during the Arvin-Tehachapi earth-
quake. This locality is about midway betMcen the "White
Wolf and Garlock faults. Four of the five chimne.vs on
the ranch house were thrown down in the direction of
the Garlock fault and it was thereupon presumed that
this fault was responsible for the shock. But further
inspection revealed two cement sacks, once resting upon
a low stone wall, thrown in the opposite direction, that
is, west. If a conclusion can be drawn from this, it is
that Branner and Davison are right.

In the 184 years since the first human record of an
earth shock in California there have been about 5000
feelable quakes each year in the California-Nevada area,
or about 2^ percent of those felt in the entire world and

• Bulletin Seis. Soc. America, vol. 5, No. 1, 1915

(137)

138

Earthquakes in Kern County, 1952

[Bull. 171

about 90 percent of those felt in the United States. 1200
were felt per year in southern California, or about one-
half of 1 percent of the world's quakes. This means that
there is an earthquake of sufficient intensity to be felt by
a person somewhere in the California-Nevada area every
hour and forty-five minutes, on an average, year in and
year out. But of the 920,000 feelable earthquakes since
the first record, only 43 can be classed as major shocks
with a Rossi-Forel intensity of VII+ or greater. The
incidence of major shocks averages only one every 4.3
years, which may offer comfort to some.

LIST OF MAJOR EARTHQUAKES, 1769-1952,
CALIFORNIA-NEVADA AREA

1769. July 28. Los Angeles region (near Olive).
"On the 28th, when the governor (Portolal and his
followers were on the Santa Ana River, four vio-
lent shocks of earthquake frightened the Indians
into a kind of prayer to the four winds, and caused
the stream to be also named Jesus de los Temblores.
Many more shocks were felt during the following
week ; yet the foreigners were delighted with the
region ..." (Bancroft, Hist. Cal., vol. 1, p. 146,
quoting from diaries of the expedition.) Both Holden
and Townley rate this as a shock of R-F YI but
Wood and Allen rate it as R-F VIII.

1790? Inyo County. Indian legends have it that a
great earthquake similar to the one of 1872 occurred
in the Owens Valley eighty years earlier. This quake
is commonly assigned an intensity of R-F X. It
would seem that a quake of this intensity would be
widely felt and hence recorded in Spanish or Mis-
sion archives but it is not.

1800. October 11 to 31. R-F VIII or IX. San Juan
Bautista.

"There were shocks of earthquake from the 11th
to the 31st of October, sometimes six in a day, the
most severe on the 18th. Friars were so terrified
that they spent the nights out of doors in the mis-
sion carts. Several cracks appeared in tlie ground,
one of considerable extent and deptli on the banks
of the Pajaro, and the adobe walls of all the build-
ings were cracked from top to bottom, and threat-
ened to fall. The natives said that such shocks were
not uncommon in that vicinity, and spoke of sub-
terranean fissures, or caverns, caused by thiem,
from which salt water had issued." (Bancroft. Hist.
Cal, vol. 1, p. 559). On November 22 at San Diego,
there was a shock of about R.F. VIII. "The earth-
quake occurred at 1:30 p.m. and the soldiers'
houses, warehouses, and the new dwelling of the vol-
unteers (in the Presidio) were considerably
cracked." (Bancroft, Hist. Cal., vol. 1, p. 654).

1812. May to December. R-F IX-X. San Juan Capis-
trano (December 8).

Forty lives lost by destruction of the mission,
where Mass was being said. Damage also at San
Gabriel. (December 21). Damage at San Fernando,
San Buenaventura, Santa Barbara, Santa Ynez,
and Purisima. Santa Barbara and Purisima Mis-
sions were completely wrecked and Santa Ynez was

damaged. A huge earthquake wave was reported at
sea which broke along the Santa Barbara coast. A
ship at Refugio was carried up a canyon and re-
turned to sea. There was no record of loss of life at
any of the missions. It has been suggested that this
quake of December 21 had its origin on a submarine
fault some miles oiTshore between Santa Barbara
and Gaviota. The reported sea-wave resembled that
of the other offshore quake of November 4, 1927
(q.v. ). 1812 was recorded in mission archives as
' ' el aiio de los temblores. ' '

1836. 1 June 9th or 10th. R-F VIII to X.

San Francisco Bay Region, possibly originating
along Hayward fault, as did the great shock of
1868. Great fissures were said to have opened at the
surface of the ground and aftershocks continued
for a month.

1838. June and July. R-F VIII. San Francisco, San
Jose, Santa Clara and Monterey.
Severe in San Francisco Harbor and damaging
at Monterey and Redwood City. This shock has
been ascribed by some writers as originating on
Hayward fault but evidence seems to point properly
to the San Andreas fault.

1852. November 9th. R-F VIII to X. Region of Ft.
Yuma.

". . . . tlie camp was violently sliaken by an
earthquake, and the shocks continued almost daily
for several months after, and were so frequent and

expected as not to excite remark the first

shock threw down a portion of Chimney Peak (20
miles NE of Ft. Yuma) and opened fissures and
cracks in the clay strata of the desert bordering
the Colorado." Active mud volcanoes, with tem-
peratures up to 170° F., were noted 40 miles south-
west of the post. (Blake, Pac. R.R. Repts, vol. 5, p.
115-116, 1856).
November 26-30. R-F IX. Southern California.

1. Eleven strong shocks at San Simeon, Los Ange-
les and San Gabriel. Felt as far south as Guay-
mas, jMexico.

2. Long sequence of shocks felt from San Luis
Obispo to the Colorado River and south to San
Diego. Thirty-mile fissure reported in Lockwood
Valley, Ventura County, near the San Andreas
and Big Pine faults.

3. Two-minute shock at San Diego, followed by
lighter ones for several days. Townley and Allen
suggest that this epicenter may have been at
Ft. Yuma with R-F IX owing to the long dura-
tion at San Diego.

1853. October 23. R-F VIII. Humboldt Bay.
Three heavy shocks. Houses were reported to have

rolled like ships at sea and a wharf sank 4 feet.

1857. January 8 and 9. R-F IX-X. Fort Tejon.

One of the three or four strongest shocks in Cali-
fornia since the advent of the white settlers. It was
strongly felt from Ft. Yuma to Sacramento but was
most violent at and near Ft. Tejon. Here all build-

Part TT]

Seismology

139

in<rs and bi": treos wero thrown down, and a fissure
20 feet wide and 40 miles long appeared, but closed
with such foroe that a ridgre 10 feet wide and sev-
eral feet hifrh was formed. Byerly says this ridge
still exists at the head of Terwilliger Valley in Los
Angeles County. Among the many things done b.v
this shock were : caved in the roof of the Ventura
mission, reversed the flow, temporarily, of the Kern
River, threw the Los Angeles River out of its bed,
formed new springs at Santa Barbara, changed part
of the course of the San Gabriel River, and caused
a great rumbling over most of the area of shock.
This earthquake without doubt had its origin along
the San Andreas fault from the Cholame Valley to
San Bernardino and the epicentral region was per-
haps in the Carrizo Plain with the Elkliorn scarp as
the surface evidence remaining.

1838. November 26. R-F VII to IX. Sau Jose-San
Francisco.

Very considerable damage to scriTctures in San
Jose, somewhat less in San Francisco where the
daily papers described it as "a violent earthquake
. . . consisted of two shocks, separated by an in-
terval of a few seconds ... at ilusical Hall, where
the Independent National Guard was having a ball,
the shock was not noticed on the dancing floor,
though the building was very much shaken" (!).

1860. March 15. R-F V-VI at Sacramento, VII at Car-
son City.

Felt as far east as Utah. A quake of large mag-
nitude but with epicenter in sparsely populated
areas, hence no reports of great damage.

1864. March 5. R-F VI to VII+. San Jose, Stockton,
Santa Rosa, San Fi-anciseo, Santa Clara, Santa
Cruz. Light at Visalia, strongest at San Fran-
cisco.

1863. October 1. R-F IX. Eureka and Fort Hum-
boldt.
"... scarcely a house in town escaped fracture
in its brickwork."
October 8. R-F IX. San Francisco-Santa Cruz.

Two strong shocks close together. Structural dam-
age in San Francisco largely confined to buildings
on made ground and service mains in it. Some As-
suring, especially in the Santa Cruz Mountains, and
some rock slides. Brick structures overthrown at
New Almaden mine. Epicenter probably on San
Andreas Rift nearb.y.

1868. September 3-28. R-F IX. Headwaters of Kern
River, Inyo County.

About a thou.sand shocks during this period, some
very severe "... tall trees swayed and even the
grass was observed to wave back and forth. Im-
mense masses of boulders and earth were detached
from the surrounding cliffs."

October 21. R-F X. Hayward. IX San Francisco.

One of California's major earthquakes. Duration

42 seconds. 49 aftershocks, some heavy, reported

to November 16. Most damage at Ha\"ivard and

San Leandro, with 30 lives lost, mostly by falling

brick. San Francisco damage again largely to struc-
tures on made ground. Epicenter along Hay^vard
fault, with horizontal .surface displacement from
San Leandro to Warm Springs (20 miles).

1871. March 2. R-F VII? Humboldt County.
Chimneys thrown down at Eureka, Petrolia,

Rohnerville, Hydesville. Duration 20 seconds. "Se-
verest for several years."

1872. March 26. R-F X. Owens Valley, Inyo County.
Commonly regarded as largest earthquake in

California in historic time. Every ma.sonry house in
Lone Pine levelled. 27 fatalities, and 60 serious in-
juries. Few frame buildings but none seriously
damaged. Ground disturbed for the 70 miles from
Haiwee to Bishop along Owens Valley fault system,
ilaximum movements: horizontal, 20 feet; vertical,
23 feet. Shock felt in all of California and Nevada,
parts of Utah and Arizona (about 125,000 square
miles were sharply affected).

1873. November 22. R-F VII to X? Del Norte
County and southern Oregon.

Felt from Portland to San Francisco but most
severe in Crescent City, with reported damage to
every brick building. Felt inland at Redding,
Yreka, and Red Bluff.

1885. April 11. R-F VIII or higher. Central Coast
Ranges.

Felt from Marysville to Ventura with probable
epicenter on San Andreas Rift between Cholame
and San Benito. A large quake but with small dam-
age owing to unpopulated area of highest intensity.

1890. February 9. R-F VII ? Southern California.
Three distinct shocks felt at Pomona, Los An-
geles, San Diego. Character of waves and duration
in above cities indicated a quake of large magnitude
originating in the San Jacinto Mountains.

1892. February 23. R-F X ? Southern California and
Lower California.

Felt from Ensenada to Visalia. Plaster fell and
walls were cracKed in San Diego area. Epicenter
probably in uninhabited region of Baja California
east of Ensenada. This is perhaps the strongest
.shock reported in the period of 1873-1906.

April 19-21. R-F IX-X. Solano County.

Extreme damage at Vacaville and Winters, es-
pecially to brick and stone structures. Fissures
formed in bed of Putah Creek. Slight damage in
San Francisco and Sacramento. Felt from Red Bluff
to Fresno and as far east as Reno. The shock of
April 21 may have been less intense than the one
on April 19, but damage was just as severe owing
to already weakened structures.

1898. April 14. R-F IX or X. Mendocino County
Coast.

^Mountain roads blocked by landslides and fallen
trees, frame houses damaged at Greenwood, fall of
chimneys and tombstones at Mendocino. Felt as far
south as San Jose. Aftershocks continued for weeks.

140

Earthquakes ix Kern County, 1952

[Bull. 171

1899. December 25. R-F IX or X. San Jacinto.

"Generally felt in .southern California and Ari-
zona. Brick and adobe structures were wrecked at
San Jacinto and Hemet. Six Indians were killed and
eight injured by collapsing adobe walls on the
Coaehella Reservation. A large-magnitude quake,
felt over an area of 100,000 square miles."

1901. March 2. R-F IX. Stone Canyon, Monterey
County.

"There were surface cracks in the ground, some
of them hundreds of feet in length. ... In some
places there was a vertical displacement of one
foot." Felt over an area of 40,000 square miles.
Epicenter probably on San Andreas rift zone north-
west of Cholame Valley.

1902. July 27, 31. R-F VIII to IX. Los Alamos,
Santa Barbara County.

Severe locally, throwing down oil tanks near
Lompoc and twisting and breaking surface oil
pipes. At least one oil well (Lompoc Oil and De-
velopment Company #1 ) was lost by casing failure.
Two more severe shocks occurred on July 31 and
completed the total damage score at Los Alamos.
Cracks, fissures and landslides contributed to the
5-day "reign of terror." Everybody left town.

190.1. January 23. R-F X? (V or VI at San Diego).

Imperial Valley.

This shock was recorded by seismographs all
over the world and was no doubt of great intensity
at its epicentral area, the uninhabited area south of
Imperial Valley in Baja California.

1906. April 18. R-F X, Central California Coast.

"The San Francisco Earthquake." Probably the
best known and certainly the most documented of
the three great shocks of California history. Dura-
tion 40 seconds in San Francisco. 270 miles of sur-
face rupture along San Andreas Rift from Fort
Bragg to San Juan. Maximum horizontal displace-
ment 21 feet near Olema. Vertical displacement
small, and at north end. Perceptible over 375,000
square miles. Total damage to San Francisco by
quake and resulting fire estimated from 350 million
to 1 billion dollars. Total casualties in San Fran-
cisco between 500 and 1,000, 300 out of the city.
All knoAvn quake effects observed on men, animals
and things. 938 ± aftershocks recorded from April
18 to June 10, 1907.

1909. October 28. R-F IX. Humboldt County.

Greatest damage at Rhonerville, M-ith all brick
and concrete structures reported damaged or de-
stroyed. Lasted 22 seconds at Eureka. Felt over
northwestern California and southwestern Oregon.
Shaken area estimated at 100,000 square miles.

1915. June 22. R-F IX. Imperial Valley.

Two violent shocks, separated by 57 minutes, af-
fected an area of over 50,000 square miles in south-
ern California, western Arizona, and northwestern
Mexico. Greatest damage ($900,000) in El Centro,
Calexico, and Mexicali. Ei)ii'entcr along (?) San

Jacinto fault near latter two towns. Six casualties
in jMexicali.

October 2. R-F X. Pleasant Valley, Nevada.

A very great earthquake with high intensity and
large magnitude. Felt from "\Va.shington to the
Mexican border and from the Pacific .shore to ]\Ion-
tana, Wyoming, Colorado, and Arizona, or an area
of 500,000 square miles. Epicenter along great
scarp which appeared suddenly at western pediment
of the Sonoma Range south of Battle ^Mountain.
Vertical displacement was 2 to 15 feet for 22 miles.
After 38 years, this scarp still looks quite fresh.
Damage slight, owing to lack of inhabitants. Very
strongly felt in northeastern California (R-F
IV-V)."

November 20. R-F IX to X. Volcano Lake, Baja
California.

Damage in Imperial Valley and Calexico. In-
tensity greatest at Volcano Lake near mouth of
Colorado River. Seismograms indicate shaken area
exceeded 120,000 square miles.

1918. April 21. R-F IX-X. San Jacinto and Ilemet,
Riverside County.

Ground cracked along San Jacinto fault, but no
evidence of displacement. Chief damage to brick
and artificial stone buildings. Xo loss of life. Felt
over southern California from Taft to Mexico and
east into Arizona. Area affected not less than 150,-
000 square miles.

1922. January 31. R-F X. Submarine, northwest of
Cape Jlendocino.

Intensity VI at Eureka. Recorded at 106 seismo-
graph stations throughout the world. Shaken area
at least 400,000 square miles. Jlagnitude probably
as great as the shock of April 18, 1906.

March 10. R-F Vlll-f . Cholame Valley, Monterey,
and San Luis Obispo Counties.
Cracks in groTind along San Andreas fault.
Chimney and house damage at Parkfield. Felt
throughout central California and as far south as
Los Angeles. Shaken area perhaps 100,000 square
miles. Recorded at 43 stations, over most of the
world.

1923. January 22. R-F IX. Submarine, near Cape
]\Iendocino coast.

Damage at Petrolia, Dyerville, Ferndale, Alton,
and other nearby towns. Recorded at 71 stations
throughout the world.

1925. June 29. R-F IX-X. Santa Barbara and west-
ward.

Nearly destroyed business district, especially
poorlj' constructed buildings on made land in lower
State Street. Felt from Watsonville through ]\Io-
jave to Santa Ana. Total area affected at least 100,-
000 square miles. Recorded throughout the world
and had an unusual number of aftershocks. Possible
epicenters were Mesa Fault (submarine extension)
aiul ((uickly triggered Santa Ynez fault, according

Part II]

Seismology

141

to Bailey Willis. Not a ^rcat shock, but of hifrh
intensity at a thickly populated area, resulting in
sevei'al deaths.

1927. November 4. R-F IX-X. Submarine, west of
Pt. Arguello, Santa Barbara County.
R-F IX near Surf with production of small sea
wave of 6-foot rise. Probably R-F VIII at Lompoc,
where chimneys fell and brick buildings damaged.
Recorded over the world as a stronger shock than
the Santa Barbara quake of 1925.

1932. June 6. R-F VIII+. Submarine, near Eureka.
Much damage at Eureka and Areata. One death
from falling chimney.

1932. December 20. R-F X. "Western Nevada near
Cedar Mountain.

As strong as the 1915 Pleasant Valley shock.
Surface rifts noted in belt of faulting 4 to 9 miles
wide and 38 miles long. Felt over Pacific states.
Little damage owing to lack of inhabitants.

1933. March 10. R-F IX. Long Beach and vicinity.
Not a great shock but as it occurred in a region

of dense settlement with many buildings of poor
construction, it ranks second onl.v to the San Fran-
cisco quake of 1906 in destructive effect. Over 100
lives were lost and monetary damage reached an
estimated 40 million dollars. Felt over a sea and
land area of about 100,000 square miles. Epicenter
just offshore along Inglewood fault.

1940. May 18. R-F X. Imperial Valley.

Eight deaths, 20 injuries. Damage to buildings,
crops, canals, and railroads over six million dollars.
Caused visible surface fault about 45 miles long
from Imperial to Volcano Lake in Baja California.
Maximum displacement was 14 feet, 10 inches, hori-
zontally; apparently no vertical displacement. Felt
over an area of 60,000 square miles in southern
California and northern Baja California.

1952. Julv 21 ; August 22. Arvin-Tehachapi. Bakers-
field.

See this volume for data.

While the eartluiuakes listed above have been very
briefly treated, it is felt that the discussion is adequate
to give the reader a fair appraisal of what is known of
historic shocks in the California-Nevada region. For
tho.se who wish to inquire further into the matter, a
list of definitive works is given below; from it this
article was largely derived.

1. Holden, Edward S. Catalogue of earthquakes on
the Pacific Coast, 1769-1897. Smithsonian ilisc.
Collections, No. 1087 (1898).

2. McAdie, Alexander. Catalogue of earthquakes on
the Pacific Coast, 1897-1906. Smithsonian Misc.
Collections, Vol. XLIX, no. 1721 (1907).

3. Townley, Sidney D., and Allen, Maxwell W. De-
scriptive catalogue of earthquakes of the Pacific
Coast of the United States, 1769 to 1928. Bull.
Seismological Soc. America, vol. 29, no. 1, Jan.
1939. (This list is the most recent and complete
catalog and offers many additions, corrections and
emendations to the two previous lists.)

4. Wood, H. 0., Allen, M. W., and Heck, N. II.
Earthquake history of the United States, Part II —
California and Nevada. U. S. Dept. Commerce,
Coast & Geodetic Survey, Serial 609, 1939.

Two important papers must be mentioned here, for
they throw mucli light on any inquiry into the field of
earthquake study. One is the Report of the State Earth-
quake Investigation Commission (Carnegie Institution
of Washington, 1908) on the San Francisco earthquake
of 1906, without doubt the most comprehensive and
detailed piece of research ever done in the field of seis-
mology. The other paper is by George D. Louderback
and is an account of the history of the University of
California seismographic stations. It was published in
the Bulletin of the Seismological Society of America
for January 1941. Here Dr. Louderback has shown that
the first earthquake-recording instruments ever used in
North America were set up simultaneously at Berkeley
and Mount Hamilton (Lick Observatory) in 1887 and
the first shock ever instrumentally recorded in the
United States was one on April 24, 1887, with an in-
tensity of R-F II.

3. SEISMIC HISTORY IN THE SAN JOAQUIN VALLEY

By C. F. Richter

The part of Kern County most affected by the earth-
quakes of 1952 has been shaken in the past about as
hard and as frequently as most sections of California.
Geologists and seismologists examining the historical
record have attributed most of this disturbance to the
great faults marginal to the area — the Sau Andreas
fault, the Garlock fault, and the major Sierra Nevada
fault. Instrumental records of recent years, in this area
as in others, show that minor shocks originate at points
rather generally peppered over the map, and only the
larger shocks can be taken as related to the principal
faults. Geological field evidence, in agreement with the
imperfect historical record, indicates that no great
earthquake is likely to have originated on the western
part of the Garlock fault in historical time.

Most of the information obtained before 1927 is non-
instrumental. The most complete account available is
that by Townley and Allen (1939), from which most
of the following list is abstracted.

1852 October 26. Strong at San Simeon, possibly re-
lated to the nest or an error in date.

1852 November 26. (See VanderHoof, Part II, 1)

1853 February 1. Violent shocks at San Simeon.
1853 June 2. Plains of the San Joaquin. Two smart

shocks. Similar shocks apparently on July 12 and Sep-
tember 2.

1857 January 9 (See VanderHoof, Part II, 1)

Several authors have considered that the shock of
1857 was larger than that of 1906 ; the present writer
prefers the opposite opinion.

1868 September 4. A strong shock apparently origi-
nating near the headwaters of the Kern River, where a
party was in camp. Numerous aftershocks, many of them
felt at Lone Pine.

1872 JIareh 26. The great Owens Valley earthquake,
felt stronglv in the San Joaquin Valley.

1882 December 19.. Shock felt at Baker.sfield and Vi-
salia.

1885 April 11. Strong shock originating in the Coast
Range, probably in Las Tablas district about 30 miles
northwest of San Luis Obispo ; more probably associated
with the Nacimiento fault than with the San Andreas
fault. Strong as far east as Visalia ; plaster cracked at
many places in the San Joaquin Valley.

1889 September 29. Strong near Bishop ; felt as far as
Bakersfield.

1890 February 13. Three light shocks felt at Tehachapi.
1890 July 24 or July 25. "Severe" at Bakersfield; felt

at Porterville.

1894 July 29. Felt from Baker.sfield to San Diego.
Minor damage (goods off .shelves, etc.) at Mojave .and
in the Los Angeles area.

1896 August 17. Plaster cracked at Hanford. Clocks
stopped at Bakersfield and Merced. Felt at Fresno and
Visalia.

1903 January 7. Alarmed persons into the streets at
Bakersfield.

1905 January 5. Felt at Bakersfield and Wasco; appar-
ently also reported at Lone Pine, Claremont and River-
side.

1905 March 18. Felt at Isabella and Wasco. Heavy at
Bakersfield, still more so at McKittrick.

1905 December 23. At Bakersfield much plaster fell,
goods were thrown off shelves, and wide cracks opened
in buildings. Much alarm. Felt at Wasco and Tejon
Ranch.

1906 April 18. San Francisco earthquake; affected
most of California. In the southern San Joaquin Valley
it was relatively light ; at Bakersfield windows and doors
rattled, and some clocks stopped. The shaking was no-
ticed at Isabella.

1908 November 4. Strong shock in the Death Valley
region, felt at least as far as Tehachapi.

1910 May 6. Strong at Bishop; rock slides in Rock
Creek. Felt as far as Bakersfield.

1915 May 28. Earthquake in the southern Sierra Ne-
vada, strong enough to record at distant seismograph
stations; sharp shock at Lone Pine and Bakersfield; felt
northwest as far as Merced ; reported at Glennville and
California Hot Springs.

1916 October 22. Strong shock centering at Tejon
Pass, felt over wide area. Probably, but not certainly,
on the San Andreas fault.

1919 February 16. A shock similar to the preceding
but centered farther west ; damage occurred at Maricopa,
and the shaking was strong at Belridge, Lebec, Grape-
vine station, and Gorman. The epicenter cannot have
been far from that of 1952, but may nevertheless have
been on the San Andreas fault.

1920 November 20. Dishes off shelves at Taft. Felt at
Maricopa.

1921 March 26. Felt at Maricopa.

1921 November 15. Slight shock at Bakersfield and
Edison.

1922 March 10. Large shock centering on the San
Andreas fault. Damage at Parkfield and Cholame. Felt
across the San Joaquin Valley and into the Sierra as
far as Springville.

1922 August 17. Strong aftershock of the preceding,
felt at least as far as Bakersfield.

1926 June 30. Strong shock in Kern River Canyon,
with rock slides; sleepers awakened at Bakersfield and
California Hot Springs; abrupt shock at Glennville.
Light shock felt at Porterville, Lindsay, Tulare, Visalia,
and as far as Pasadena and San Luis Obispo. This
incomplete information suggests an epicenter near that
of 1952, July 29.

Beginning with 1927, seismograms from the Southern
California stations are available, but for the first few
years epicenters are located approximately only. The
following data are from files at Pasadena.

1927 July 8. At Bakersfield, dishes rattled heavily;
felt by motorists ; one parked car shifted. Felt sharply
in Kern River oil fields ; one abandoned well returned
to production. Noticed at Fellows but not at McKittrick.

1927 September 17. Damage at Bishop ; shock felt in
the Sierra Nevada and San Joaquin Valley, and as far
as Palmdale.

1929 March 12. Felt from Delano to Ventura; rather
generally noticed in the southern San Joaquin Valley.

(143)

144

Earthquakes in Kern County, 1952

[Bull. 171

s ?=

Part III

Seismology

145

Tnhle 1.

I nstrumentaUy located epicenters (platted in fig. I), 19Si and I'.tS'i-

118' lo'-lZO" 00' W.

-June 30, 1932. Limits Lat. 3|° J,5' y to .W° /,.'>' \. Long.

Date

1932— Jan. 7

Feb. 14

Apr, 19

June 22

July 25.

1934— Apr. 30

May 6..

July 6

July 12

Aug. 25-

Sept. 27

Oct. 13

Nov. 16

Dec. 5

Dec. 21

1935 — Jan. 23..

Mar. 4-

Mar. 5

Mar. 17

Mar. 18

Apr. 13

May 10

6 shocks + 35 others prior
to midnight May 18/19

June 9

June 11

June 18.

July 6

July 7

Sept. 10

Sept. 22

Oct. 27

Nov. 23

1936— Feb. 3..

May 3

May 6

May 30

Aug. 4-

Aug. 20

Sept. 26.

Oct. 5 -.

Oct. 9

Nov. 28

Nov. 30

Dec. 22

1937— Jan. 19

Apr. 22.

June 8

Oct. 4

Nov. 27

Dec. 11...

Dec. 19

1938— July 1

Oct. 15

Dec. 4...

1939— Feb. 23..

Feb. 23

Feb. 23.

Mar. 7...

Apr. 14

May 7..

June 20

July 21..

July 24.

Aug. 19

Oct. 9

Oct. 25...

1940— Jan. 18

July 12.

July 29

Aug. 6...

Oct. 23

Nov. 17

1941— Jan. 20

Jan. 23

Feb. 9

Lat.
Deg. Min.

34
34
35
34
34
35
35
35
35
35
34
34
35
35
34
35
35
34
35
35
35
35

35.7

34.9

45
58
41
55
53
15
40
15
05
34
59
49
00
06
59
27
01
59
22
22
00
42

Long.
Deg. Min.

118
119
118
119
119
119
119
118
118
119
118
119
118
119
118
119
118
118
118
118
118
118

119

118

119

118

119

118

119

118

119

118

119

118

119

118

119

118

119

118

119

118.3

119

118

119

119.1

40
00
28
05
00
10
18
15
40
51
35
00
53
00
35
15
23
35
50
50
53
22

50
22
22
22
22
59
10
09
53
17
45
22
22
08
53
53
50
51
50
00
20
15
22
08
50
53
22
22
26
00
35
08
30
01
01
00
00
00
00
00
00
03
22
18
00
09

16
16
59
35
05
13
00
30
15
12

Mag.

2.5

3.5±

3.5

2.5

5
5

2.5

3.5

2.5

3.5

2.5

4.6

4.8

3.5

2.5

4.4

2.5

3.5

2.5

3.5

2.5

Quality

C
C
B
C
C
C
B
C
C
B
B
B
C
C
B
C
B
C
B
B
C
B

C
B
C
B
C
C
C
C
C
C

c
c
c

B
C
C
B
C
B
B
C
C
B
C
C
B
C
C
C
B
C
C
C
A
B
A
B
B
B
B
C
A
C
C
C
C
C
B

C
C
B
B
C
C
C
C
C

Date

1941— Feb. 21.

Mar. 13.

Mar. 30.

June 4. .

Sept. 21.

Sept. 21,

Sept. 29
1942— Aug. 10.

Dec. 5..
1943— Jan. 15.

Feb. 17.

May 19.

Oct. 7..
1944— Jan. 21.

Jan. 21.

Jan. 22-

Jan. 26-

Jan. 27-

Jan. 28.

Jan. 30.

Jan. 31.

Jan. 31-

Feb. 3..

Feb. 23.

May 31.

July 26.

Sept. 30
1945— Feb. 5..

Mar. 15,

Mar. 21,

June 16.

July 21.

July 24.

Sept. 3.

Nov. 14,

Nov. 30
1946— Jan. 17.

Feb. 13.

Feb. 15.

June 5. .

June 5..

July 23.

July 25.

Aug. 20-

Aug. 20.

Nov. 5-.

Nov. 25,

Dec. 29-
1947— Feb. 1 - -

Feb. I.-
Feb. 3..

Feb. 6..

Feb. 7_-

Feb. 9..

Feb. 10.

Feb. 12.

Feb. 17.

Feb. 25.

Mar. 18.

July 17.

Sept. 18,

Oct. 19.

Oct. 27.

Nov. 26.
1948— Feb. 5..

Mar. 14.

Mar. 19.

Mar. 20.

Mar. 23.

Apr. 3. .

Apr. 20.

May 6..

May 28.

May 31.

July 20.

Lat.
Deg. Min.

35
35
35
35
34
34
34
34
34
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
34
35
34
35
35
34
34
35
35
35
34
34
35
35
35
35
35
34
35
35
35
35
34
35
35
34
35
35
35
35
35
35
35
35
34
34
34
35
34
35
35
35
35
35
35
34
35
34
35
35
35

07-

Long.
Deg. Min.

22
18
17
18
56
56
56
56
56
25
03
00
56
26
56
51
51
55
55
55
55
55
55
55
55
55
55
30
47
47
20
00
53
00
53
55
53
57
15
55
12
58
12
38
21
21
05
04
07
25
25
00
03
13
21
21
49
42
43
43
45
45
25
44
05
43
20
05
38
55
24
05
29
25
25
25
01
58
25
30
30
58

Mag.

3.5

5.2

2.2

3.4
2.7

2.9
2.6
2.7
2.6
3.2
3.0
3.5
2.4
3.0
2.4
1
0

8
7
8
2.9
2.8
2.9
2.6
2.9
3
3
3
2
2
3

2
3
4
9
7
0
2.9
3.0

Quality

B
C
C
C
A
B
B
B
C
C

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

B
B

c
c
c

B
C

B
C
C
A
B
B
C
C
C
C
C
B
B
B
C
C
C
C
C
C
C
C
C
C
C
C
C

c
c
c
c
c
c
c
c
c
c
c
c
c
c

146

Earthquakes in Kern County, 1952

[Bull. 171

Tnhle 1.

Instnimentnlly Uicnied epirniirra (platted in fig. 1), 1932 mid i:i3', — .Jiair .W. 19.',2. Liniitx Trnt. S!,° Ji5' IS! to 35° J,.5' N, Long.

US" IS'-ISO" 00' W.— Continued.

Date

Lat.
Deg. Min.

Long.
Deg. Min.

Mag.

Quality

Date

Lat.
Deg. Min.

Long.
Deg. Min.

Mag.

Quality

1948— Sept. 16

Sept. 16

Sept 28

35 00
35 00
35 30
35 01

34 45

35 40

34 59

35 02

34 47

35 03
35 12

34 58

35 31
35 05
35 45

34 50

35 23
35 43
35 44
35 41
35 20
35 13

34 56

35 29
35 31

118 30
118 30

118 58

119 02
118 59

118 20

119 12
119 00
119 08
118 53
118 46
118 23
118 21

118 20

119 37
119 00
119 15
118 15
118 20
118 19
118 50

118 45

119 19
118 49
118 49

2.7
3.0
4.2
2.8
3.0
3.1
3.0
3.2
3.0
2.8
2.7
2.6
2.9
3.0
4.6
2.5
3.6
3.6
3.2
3.0
2.6
2.5
2.8
2.9
3.0

C
C
C
D
B
C
C
B
C
C
C
B
B
B
C
C
C
C
B
C
C
B
C
C
B

1950— Dec. 7

35 31
35 30
35 03
35 31

34 54

35 01
35 01
35 15
35 05

34 55

35 45
35 18
34 56

34 55

35 20
35 15
35 15
34 45

34 58

35 16
35.0

34 49

35 00
35 38
34 55

118 49

119 10
118 52
118 52
118 30

118 56

119 04
119 39
119 00
118 32
118 56

118 55

119 02
119 30
119 07
119 07
118 55

118 38

119 08
119.2

119 04
119 15
118 18
118 50

2.6

2.4

4.4

2.6

2.5

3.2

3.6

3.2

2.6

3.5

2.6

2.9

3.4

3.8

2.9

3.4

2.8

3.1

2.9

2.5=t

3.0

2.8

3.1

2.7

Dec. 10

Dec. 14__

Oct 27

Dec. 15

1949 Jan 22

Dec. 17-- ---

Mar 26

1951— Jan. 19 GCT

Apr 20

Mar. 26

May 13 -

July 14

May 29

1950— Feb. 10

July 27

Aug. 18

Mar 18

Oct. 28 -.-

Mar 23

Nov. 17.-- ---

Nov. 25 -

June 30

Dec. 15. -.- --

July 11

Dec. 15 -

Aug 10

Dec. 26.- --

Dec. 28

1952— Jan. 1

Nov 1

Mar. 7 ---

Nov 3

Mar. 16

Nov 28

Apr. 13

Dec 6

Apr. 24 --

Dec 7

June 14-_ -

The writer investigated this small earthquake in the
field. Strongest apparent intensity appeared to be in the
vicinity of Old River and Panama, where small objects
were moved and old frame structures were slightly dam-
aged. Seismograms at Santa Barbara and Pasadena
indicate distances of 92 and 125 kilometers from those
stations; these distances are within a few kilometers the
same as for the major earthquake of 1952.

19.30 October 30. Seismograms indicate a minor shock
near Bakersfield.

1931 April 21. Shock similar to the preceding. •

For 1932, and for 1934 to date, a bulletin on local
earth(iuakes in southern California was issued from the
Seismological Laboratory.

Table 1 is extracted from this bulletin, listing shocks
in the area Lat. 34° 45'-35° 45' N. and Long. 118° 15'-
120° 00' W. The epicenters are platted in figure 1.
Through 1950, records at Pasadena were kept in Pacific
Standard Time, with no attention to the vagaries of
daylight-saving time. This may occasionally affect the
date of shocks near midnight when compared with otlier
listing. Beginning in 1951, all records were kept in the
universally standard Greenwich Civil Time (sometimes
denoted G.M.T., but beginning the day at midnight) ;
this is eight hours faster than Pacific Standard Time,
or seven hours faster than the corresponding daylight-
saving time. Shocks during the late afternoon according
to either local time will hence appear as of the following
day G.C.T.

In the earlier part of table 1, magnitudes are given
to the nearest half unit ; beginning 1943 they are given
to the tenth. B indicates location believed trustworthy
within about 5 kilometers (3 miles), C within about 15
kilometers (10 miles) ; A indicates that the shock is
exceptionally well recorded and specially studied.

On July 25, 1932, a shock of magnitude 4.5 originated
in the upper Kern River district. The epicenter was
placed instrumentally (quality C) at 35° 48' N 118°

32' W, northwest of Kernville near the trace of the
Kern Canyon fault. The shock was sharp, but caused no
reported damage in the central area (but chimneys were
reported cracked at Springville) ; it was felt across the
Sierra Nevada from Owens Vallev to San Joaquin
Valley.

Table and map do not include the Walker Pass shocks
of 1946. The principal shock of that group occurred on
March 15, 1946; epicenter 35° 43' N 118° 02' W, magni-
tude 6j. The shock was felt over much of southern Cali-
fornia. Weak structures were damaged at Weldon, Onyx,
and some more distant points. Rock slides in Sand Can-
yon damaged the cover of the Los Angeles aqueduct.
The epicenter named is that assigned by Chakrabarty
and Riehter (1949). This was based on the time-dis-
tance curves then being used as standard in southern
California, but since revised. Mr. G. G. Shor finds that
applying the later revision will not displace the epicen-
ter more than a few kilometers.

Aftershocks of the Walker Pass earthquake tended to
spread geographicall.v in time ; especiall.v southwestward,
into the Kern River area. One on June 5, 1946, included
in the Chakrabarty-Richter study, has been placed by
Mr. Shor at a revised location Lat. 35° 39' N., Long.
118° 21' W.

The shocks in 1939 and 1941 at Lat. 34° 52'-54' N., Long.
118° 56'-n9° 01' W. are of particular consequence. They
were included by Gutenberg (1943, 1944) in studies estab-
lishing standard travel times for the area. Reinterpreta-
tion since 1949 will not materially alter these epicenters,
which are in the Wheeler Ridge block between the San
Andreas and White Wolf faults. The times of these
shocks were directly compared by Gutenberg with those
of the major earthquake of 1952, in order to derive a
preliminary epicenter for the latter (Lat. 35° 00' N.,
Long. 119° 00', only slightly modified since) ; the relative
placing of these shocks to that epicenter is therefore
unusuall}' precise.

4. SEISMOGRAPH DEVELOPMENT IN CALIFORNIA

By Hugo Benioff

ABSTRACT

A number of new forms of seisiiio!;ra]ihs liavc liccn dcveloijcd in
California. These iucUule tlie torsion seismograph, the variable
reluctance transihuer electromagnetic pendulum seismograph, the
electromagnetic linear strain seismograph and the fused quartz
secular strain page.

As California is the most seismically active state of
the Union, it is not surprising that the development of
seismographs has been prosecuted vigorously here. How-
ever, up to about 1923, seismograi)hs operating in Cali-
fornia were few in number and for the most part of old
or obsolete types. The impetus of the new program of
development was given by II. 0. Wood (1916), who was
the first to point out that for the study of seismicity of
a region such as California, a coordinated network of
stations is required in which each station is provided
with accurate time and seismographs of special char-
acteristics. Although at that time there existed in Japan
a large number of stations in a relatively small area,
they were not provided with sufficiently accurate record-
ing-drum drives and inter-station time, and the seismo-
graphs were of inadequate magnification to record the
high frequencies observed in local earthcjuakes.

The first instrument to ajipear on tlie new program
was the torsion seismograph, invented by Dr. J. A.
Anderson and developed jointly b.y him and 11. 0. "Wood
(1925). Essentially, it consists of a horizontal pendulum
in the form of a small copper mass eccentrically mounted
on a vertical taut wire suspension as shown in figure 1.
Damping of the pendulum motion is provided by the
reaction of eddy currents generated in the mass with the
field of a permanent magnet in which the mass is im-
mersed. Horizontal vibration of the ground results in
angular vibration of the pendulum mass about the sus-
pension. A small mirror attached to the mass serves to
deflect the recording light beam which comes to a point
focus on the sensitive emulsion of a paper or film
wrapped around the recording drum. For recording
rapid earth movements, the pendulum mass is con-

SUSPENSION

PENDULUM MASS

PENDULUM RESPONSE

MIRROR

CYLINDRICAL LENS

LIGHT SOURCE

DAMPING MAGNET-

REFLECTING PRISM
CYUNDRICAL LENS
RECORDING DRUM

PHOTOGRAPHIC PAPER

Figure 1. Schematic representation of torsion seismograph.

structed in the form of a small cylinder, 2 millimeters in
diameter and 25 millimeters long. The free period of
vibration of the pendtdum rotating about its suspension
is 0.8 second. With this instrument the magnification,
defined as the ratio of light spot displacement to ground
displacement, has a maximum value of 2800.

For recording the slower wave movements which are
generallj- observed in distant earthquakes the peiululum
mass is built in the form of a rectangular plate of copper
with dimensions approximately 25x8x1 millimeters.
This pendulum has a free period of 6 seconds and a
maximum magnification of 800.

The magnifications of these instruments were too high
for recording the principal ground movements in large,
nearby earthquakes and consequently a modified form
of the torsion seismograph was developed for these move-
ments by Dr. Sinclair Smith of the Mount Wilson Ob-
servatory staff. In this strong motion seismograph, the
pendulum was made up of two masses of unequal size,
mounted at opposite ends of a horizontal bar supported
by a vertical torsion suspension through its center. This
instrument has a period of 10 seconds and a maximum
magnification of 4. It wrote satisfactory seismograms of
the Long Beach earthquake of 1933 and the Kern County
shock of 1952.

However, for most routine studies of local earthquakes
the maximum obtainable magnification of the torsion
seismograph was inade(|uate. In addition, a satisfactory
instrument of this type for recording the vertical com-
ponent of the ground motion was never made. To meet
these limitations, a new form of electromagnetic pendu-
lum seismograph was developed in 1931 (Benioff, 1932).
In this instrument the movement of the pendulum, gen-
erates electric power by means of a variable reluctance
electromagnetic transducer. Recording is accomplished
with a galvanometric photographic system. Earlier forms
of electromagnetic instruments used moving conductor
transducers and were constructed with long periods and
relatively low magnification. With the magnetic mate-
rials available before 1931 it was not possible to con-
struct instruments of the moving conductor type having
sufficiently short periods and high magnifications for an
adequate study either of local earthquakes or of the short
period waves of distant earthquakes. The variable reluc-
tance transducer represents an embodiment of the tele-
phone receiver principle in which a permanent magnet
supplies magnetic flux through an associated armature
in series with one or more air gaps. In the latest model
(fig. 2) movement of the seismometer pendulum varies
the lengths of four air gaps in such a way that for a
given direction of movement of the pendulum, two of the
gaps increase in length while the other two decrease. The
resulting changes in flux through the armatures generate
emfs in the output coils surrounding them. In order to
produce a large electrical output without recourse to
amplifiers, the pendulum mass was made large — 100 kilo-
grams. In the vertical component instrument the mass
is supported by a helical spring, as shown in figure 2,
which in later models is made of an Elinvar-type alloy
having a low temperature coefficient of ela.sticity. Six

(147)

148

Earthquakes in Kern County, 1952

[Bull. 171

GUIDE RIBBON

SPRING

-INERTIA

REACTOR

AIR GAPS

-COIL
-MOVING SECTION OF TRANSDUCER
-STATIONARY SECTION OF TRANSDUCER

SCHEMATIC SECTION. VERTICAL COMPONENT

FlQUKE 2.

steel ribbons stretched radiallj' between the cj'lindrical
mass and the three steel supports of the instrument serve
to constrain the movement of the pendulum to the verti-
cal direction only.

In the original form, damping was provided by a dash-
pot mechanism in which a perforated disc attached to
the pendulum moved in a cylinder containing oil. Later
in 1932 (Benioil', 1934) the transducer design was modi-
fied (fig. 2) to increase the efficiency to the point where
damping of the pendulum was derived solely from the
reaction of the output currents. The efficiency of the seis-
mograph was thus raised to the maximum po.ssible value.
Referring to the cut-away transducer drawing (fig. 3),
M is the magnet in the form of a 3-inch-square plate,
f inch thick ; B is the flux distributing armature orig-
inally formed of laminations of silicon steel and later of
nickel steel alloy; G, G are the air gaps; A, A are the
laminated alloy armatures around which are wound the
coils C, C. The portion of the structure including the
magnet and distributing armatures B is attached to the
frame of the instrument. The rest of the transducer
structure moves with the pendulum. In addition to a
greatly increased efficiency, this transducer also provides
a negative restoring-force for overcoming approximately
nine-tenths of the positive restoring-force of the spring.
The whole seismometer can thus be made very much more
rugged than would be possible without the negative re-
storing-force. Moreover, since the gaps are large (2 milli-
meters) close manufacturing tolerances are not required.
In the latest model having an alloy spring, this pendu-
lum remains stable and in operating condition over a
temperature range of 55 degrees centigrade. The power

C - COILS

M - MAGNETS

G - AIR GAPS

A - MOVING ARMATURES

B - FLUX DISTRIBUTING LAMINATIONS

VARIABLE RELUCTANCE TRANSDUCER

H. PENIOFF - 1932
Figure 3. Cut-away drawing of viiriable reluctance transducer.

TV"

Figure 4. ^'erti(•aI component variable reluctance
elect ronuiftuetic seismometer.

Part TT]

Seismology

149

Figure

Horizontal comiuinent variable reluctance
electromagnetic seismometer.

output of this seismometer, derived solely from the
energy of the seismic waves, is sufficient to operate two
galvanometers simultaneously. The transducer is pro-
vided with eight identical coils. In the standard form,
four of the coils are connected in parallel to form a
generator of 31 ohms resistance for operation of an 0.2
second period galvanometer and the other four are con-
nected in series to form a 500 ohm generator for driving
a 90 second period galvanometer. The latest model of the
vertical component instrument as manufactured by the
Geotechnical Corporation, Dallas, is shown in figure 4.

A similar design was developed for the horizontal com-
ponent instrument shown in figure 5. In this component
the steady mass is supported by two of the six constrain-
ing ribbons. Restoring-force is provided in part by grav-
ity and in part b3' tension of the ribbons. In other re-
spects the electrical and mechanical characteristics of
the horizontal seismometer are identical with those of
the vertical component. Both are operated with a free
period of 1 second and with critical damping. With the
two standard galvanometers these seismographs have re-
corded waves ranging in period from ^ second in the
case of small local earthquakes to 4 minutes in the sur-
face waves of the Assam earthquake of August 15, 1950.
The maximum effective magnification of these instru-
ments is limited solely by the ground unrest, which is
present everywhere on earth. In regions where the unrest
is small, the maximum useful magnification approaches
500,000 for the short period galvanometer combination.

Another new type of seismograph was developed at
the Seismological Laboratorj^ in 1931 (Benioff, 1935).
Up to this time all exi.sting seismographs were of the
pendulum tj'pe in which the response is derived from
the relative motion of the pendulum mass and the vi-
brating ground. In this new form, known as the linear
strain seismograph, the response is derived from the
actual strain or distortion of the ground produced by
the seismic waves. This strain is brought about as a
result of the finite speed of propagation of seismic waves
so that the phase of motion at a given point is different
from that at another point along the line of propaga-

FlGURE 6.

Schematic representation of linear
strain seismometer.

tion. In the original form (figure 6 and figure 7) the
instrument consisted of two steel piers set into the rock
at points 20 meters apart. A two-inch iron pipe rigidly
attached to one pier extends to within a short distance
of the other pier. The pipe is suspended by 12 wire
supports which are longitudinally compliant and rela-
tively rigid in the transverse direction (figure 8). "When
a seismic wave traverses the site of the seismometer the
two piers alternately approach and recede from each
other. The free end of the pipe is thus displaced to and
fro relative to the adjacent pier and this relative motion
serves to actuate a variable-reluctance transducer similar
to the one previously described for the pendulum instru-
ment. The transducer output power is recorded galva-
uometrically as in the pendulum seismographs described
above. Since the response of this instrument is derived

Figure 7. Original tli/ttrumagnclit liiii:a

. lii ' meter.

loO

Earthquakes in Kern County, 1952

[Bull. 171

Figure 8. Transducer end of electromagnetic linear strain seis-
mometer showing pier, transducer and one o£ the supports for the
indicator tnlie.

l''l(il'KK y. \'ertical <-onipon*'iit \atiable capacity seismometer.

from groiiud strain rather than displacement, as is the
case with the pendulitm seismographs, its directional and
frequency characteristics differ radically from those of
the pendulum instruments. Observations made with this
instrument taken by themselves and in combination with
those of the peiultdum instruments, provide information
concerning seismic waves which cannot be had from
pendulum instrunumts alone. The seismogram of the
Kamchatka earthquake of November 4, 1952 written by
this instrument with a recording galvanometer of 3
minutes period, contained waves of 51 minutes period
— much longer than any waves that have been observed
hitherto. The effectiveness of this instrument for very
long period strains such as the secular strains which
generate eartlujuakcs, is limited by the thermal expan-
sion characteristic's of the indicator pipe. Thus with the
steel pipe, changes of temperature of 1 degree centigrade
produce movements of the free end corresponding to
strains in the earth of 10"^. In an attempt to measure
secular strains and also tidal strains produced by the

sun and moon, a modified form of the linear strain in-
strument is being set up in a tunnel situated in the
mountains north of Glendora. A tunnel such as this
should exhibit very small temperature variations. More-
over, in order further to reduce the thermal response of
the instrument, the indicator pipe is constructed of fused
quartz — a substance having a thermal expansion of oidy
f^ve parts in ten million per degree centrigrade. The
instrument is nearing completion at the time of this
writing. Should the preliminary experiments indicate
that this instrument operates in accordance with expec-
tations, it is hoped that a large number of instruments
of this kind can be distributed throughout the state.
With such a network it should be possible to determine
the nature of the strain patterns which generate our
earthquakes and from these learn something as to the
origin of the forces which produce strains. Moreover,
given enough time, possibly one or two centuries, a study
of the strain pattern variations in relation to the se-
(juence of earthquakes may provide a sufficient basis
for approximate predictions of the times and locations
of future earth(iuakes.

Another type of pendulum seismograph has been
developed by the writer primarily for operation of vis-
ible writing recorders and magnetic tape recorders. This
instrument is provided with a transducer of the variable
discriminator type for operation with a high frequency
oscillator of constant frequency (figure 9). The rec-
tangular mass is positioned between two sets of fixed
plates to form two condensers of equal capacity when
the pendulum is in the rest position. The two condensers
are each shunted by identical inductances. The two
tuned circiuts thus formed have the same resonant fre-
quency. The inductances are coupled to the output cir-
cuit of a crystal oscillator operating at a frequency of
5.35 megacycles. When the pendulum is in its rest posi-
tion the two circuits are each detiuied 50 ke from the
ciystal freqitency at which point they each have cur-
rents approximately 0.7 times their resonant value. Out-
])uts from the two tuned circuits are rectified by two

FlGtiHlo 10. Short period galvanometer recorder using
photographic paper.

Part II]

Seismology

151

germanium crystal diodes. The difTereiiee between the
two rei'tifii'd outputs is proportional to the movement
of the pendulum. For recording directly with a galva-
nometer the diodes are connected through two high re-
sistances to the galvanometer and a large series capaci-
tor. The capacitor thus serves to eliminate the slow
current drifts. For operation of other electronic devices,
the outpiits go to push-pull amplifiers. The amplifiers in
turn may serve to operate visible writing and or mag-
netic tape recorders. In the latter instrument, in use
at the Seismologieal Laboratory in Pasadena, record-
ing is effected at a tape speed of ^ mm/sec. When played
back at the normal 15 inches/see, the seismic frequen-
cies are accelerated approximately 750 times and are
transformed into the audio range of frequencies in which
form they can be analysed with audio frequency instru-
ments.

Galvanometer Recorders. The galvanometer recorders
use drums which accommodate photographic paper sheets
30 X 90 cm, or 90 cm lengths of 35 mm film. For local
earthquakes the paper recorders operate with a writing

speed of 1.0 mm/sec and the film recorders 0.25nim/sec.
Figure 10 shows one of the paper recorders operating
with a short period (0.20 sec.) galvanometer. This drum
was designed by Howell and Sherburne of Pasadena and
has come to be known as the Henson drum. (Henson was
one of the early manufacturers of seismographs de-
scribed in this paper.) For each revolution the drum
advances axially 2| mm by means of a screw located
within the axis of the instrument. One standard sheet
is thus covered in 24 hours with successive recording
lines 15 minutes apart. The film recorders operate at
one-fourth the speed of the paper recorders and the line
spacing is also reduced. However, the increased resolu-
tion of film as contrasted with paper more than offsets
the effects of the slower speed. In the early daj's the
power line was not controlled in frequency and conse-
quently in order to rotate the drums at a sufficiently
uniform rate a tuning fork controlled drive was devel-
oped. However, at present most installations are driven
by small synchronous motors operated by the 60 cycle
power line.

5. SEISMOGRAPH STATIONS IN CALIFORNIA
By B. Gutenberg •

ABSTRACT

A short history of seisniosriiph stntions in California is given.
Stations in neighliorins states which lonlrihuted information to
the study of the Kern County earth(;uakes in inr>2 are listed.
Detailed information is given for stations in California which re-
corded the shocks in lit52 ; this includes a list of installations with
portable instruments in the epicentral area from July 21 to Xo-
vember 13, 1952.

The first instruments in the United States to record
earthquakes seem to have been installed at Berkeley and
at Lick Observatory in 1887 (Louderbac-k, 1942)". The
equipment at each station included two horizontal Ewinp:
seismographs and one seismograph to record the vertical
motion. The recording was started by the earthcjuake,
and the three traces were recorded on the same rotating
disk. Minute marks were made by a clock. Similar in-
struments were operating temporarily at several other
locations in the same area. The San Francisco earth-
quake of 1906 was recorded at Berkeley. Oakland, Yount-
ville, Alameda, San Jose, Los Gatos, Lick Observatory
in California, and at Carson City, Nevada ; most of the.se
instruments had magnifications of about 4. Records were
discussed and reproduced by Reid (1910). In 1910 in-
struments with higher magnification were installed at
Berkeley and Lick Observatory and later better instru-
ments and other stations were added with Berkeley as
central station.

The need for a network of seismic stations in southern
California was emphasized by Wood (1916) and, as a
result, a network with Pa.sadena as central station was
inaugurated jointly by the Carnegie Institution of
Washington and the California Institute of Technology
in 1923 (Day, 1938). In 1936, the Carnegie Institution
transferred their part to the California Institute of
Technology which has maintained and expanded the
network since.

The two networks provide the bulk of the stations in
California. Another of the oldest stations is that inaugu-
rated in 1909 by the I'niversity of Santa Clara at the
Ricard Observatory. The U. S. Coast and Geodetic Sur-
vey, recognizing the need for additional data about
earthquakes in California, has installed two stations
there, one near ITviah, the other at Shasta in cooperation
with the Bureau of Reclamation. Records of the Shasta
station are now measured at Berkeley. In addition, the
U. S. Coast and Geodetic Survey is operating many
strong-motion instruments throughout California (see
Part II, 12).

For location and study of earthquakes in California,
records or reports of a number of stations in neighboring
states are frequently used. Among them are the follow-
ing which have made available records of the Kern
County shocks for the present investigation : the Domin-
ion Observatory station at Victoria, B. C, with three
auxiliary stations; the station of the Division of Seis-
mology at the University of Washington at Seattle; the
station at the Department of Physics, Oregon State Col-
lege at Corvallis, Oregon; the seismological station at
Mt. St. Michael's, Spokane, Washington; the following

• Manuscript received for publication July 13, 1953.

five stations, operated by the U. S. Coast and Geodetic
Survey, partly in cooperation with local institutions (as
indicated) : Boulder City, Nevada (Bureau of Reclama-
tion), Bozeman, ^lontaiia (Jlontana State College),
Butte, Montana (^Montana School of Klines), Hungry
Horse, Jlontana (Bureau of Reclamation), Tucson, Ari-
zona ; furthermore the following independent stations :
the station at the University of Nevacla, Reno, Nevada
(records are measured at Berkeley) ; the seismological
station at Regis College, Denver, Colorado; the seismo-
logical observatory of the Texas Technical College at
Lubbock, Texas ; several seismological .stations in Mex-
ico, operated by the Institnto de Geofisica, L^niversidad
Nacional de ]\Iexico.

The following data on California stations which fur-
nished records for the investigation of the Kern County
shocks include the location of the instruments, and (for
general information only) the main characteristics of the
instruments. Cooperating agencies and institutions are
given in parentheses. However, correspondence should
be directed to the respective central stations. (There
have been many changes and additions since the time of
writing.)

Abbreviations used :
N = North latitude
W = West longitude
H = elevation in meters

BS = Benioff seismograph with short-period galvanom-
eter; period of pendulum about 1 second, galva-
nometer period about 5 second.

BL = Benioff seismograph with long-period galvanome-
ter; same pendulum as preceding, galvanometer
period of the order of 1 minute.

TS = standard Wood-Anderson torsion seismograph,
period 0.8 seconds; maximum magnification about
2800 (for waves with periods of less than | sec-
ond).

TL = similar instrument with period of about 6 sec-
onds; maximum magnification about 800 (for
waves with periods of less than 5 seconds).

G = Galitzin seismograph ; period of pendulum and
galvanometer roughly 12 seconds.

S = Sprengnether seismograph ; periods of instru-
ments and galvanometer about 2 seconds ; maxi-
mum magnification (for periods of about IJ sec-
onds) about 3500 for horizontal components and
amount 5000 for vertical components.

V = order of magnitude of maximum magnification for
continuous sinusoidal waves ; these values change
considerably with time and should not be used for
calculations without consulting the respective .cen-
tral stations. Where pos.sible, the approximate
ground period (or range of periods) to which V
applies is added in parenthesis. Recording on film
is indicated by VF which then refers to the rec-
ord as viewed on the screen of a standard pro-
jector with 8x magnification.

(153)

154

Earthquakes in Kern County. 1952

[Bull. 171

Figure 1. Map showing locations of permanent anil semi-permanent seismological stations in California during 1952.

Part TTl

Seismology

155

Z, RW, NS refer to vcrtii-al, east-vvost and iiorth-sdiitli
I'ompoiients respoetively ; the orientation of tlic
liorizontal instruments may deviate by ±10° from
the given direetions.

A. Instruments of the Pasailina group of stations:

Pasadena, Seismologiral Lahoialory ; N—3.',''08.n' ; W=l IS°10.S' ;
H = 2!t5. Mailing luUlro.ss for all stations of this group: 220
North San Rafael Avenue, Pasadena 2, California.
BS, Z.EW.NS; V=30.000 (0.2 sec.)
BI.., Z.EW.NS; V = 2,000 (1 sec.)

Benioff capacity seismograph, Z,NS; V=8,000 (1 sec.)
Beniotf strain seismograph, NS ; V=300 (i to 20 sec.) ; I'^W ;

V=100±: (} to 20 sec.)
TS, EW and NS TL, EW and NS.

Strong-motion seismograph, EW,NS ; V=4 (0 to 5 sec.)
Mount Wilson: N=3!,° 1S.5'; W = US' 03.V ; H = 1742
(Mount Wilson Observatory, Carnegie Institution of Wash-
ington ) .
BS, Z ; V = 30,000 (0.2 sec.)

S.'J.fi'; 1\' :

22M'\ H = 250 (City of

Riverside : .Y = ;
Riverside) .
BS, Z; V = .30,000 (0.2 sec). TS EW and NS
J'alomar; N - 33° 21.S'; W = 116'> 5J.6'; H = 1700 ( Palomar
Oliservatorv, California Institute of Technology).
BS, Z; V = 30,000 (0.2 sec). EW and NS ; VF = 30,000 (0.2
sec)
La JoUa: N = 33° 51.8'; W = in" 15.2' \ H = 8 (Scripps Insti-
tution of Oceanography, University of California).
TS, NS ; V = 2,800. Discontinued July ,30, 1952.
Santa ISnibara : N = 3J,° 26.5'; W = 119° J,2.9' ; H = 100 (Santa
Barbara JIuseum of Natural History).
BS, Z: VF = .3000± (0.2 sec)
TS, EW and NS ; VF = 2800 (0 to } sec.)

TS, NS ; V = 2800 (0 to i sec). Discontinued on Dec. 23, 1052.
China Lake: N = 35° jO.O' : ^ = 117° 35.8'; H = 71U5 (Naval
Ordnance Test Station).
BS, Z; V = .50,000 ± (0.3 sec). BS, EW, NS ; VF = .30,000 ±

(0.2 sec.)
BL, Z; VF = 10,000+ (1 sec)

Haiuee; N = SG" 08.2': W - 1 17° ■57.9'; H = 1100 (Bureau of
Water and Power, City of I^os Angeles).
BS, Z ; V = fi,00O± ( 0.2 sec )
TS, EW and NS.

Tinemaha: N = 37° 03.3'; W = 118° 13.T; H = 1180 (Bureau of
Water and Power. City of Los Angeles).
BS, Z; V = 30,000 (0.2 .sec)

BL, Z; V = lOOOi (1 sec). NS ; V = 2000± (1 sec)
TS, EW and NS.

Dalton; .V = ,■?.', ° 10.2'; W = 117° 1',.0'; H = .523 (Los Angeles
County Flood Control District).
BS, Z; VF = ,30,000± (0.2 sec) ; VF = 2,000± (0.2 sec)

Big Bear; N = 3J,° 11,.S' ; W = 116° 51,.8' ; H = 2060 (Big Bear
Lake Elementary School, Big Bear LaUe, California).
BS, Z; V = 30,000± (0.2 sec.)

Barrett: X = 32° -',0.8'; W = llli' 1,0.3'; H = 510 (Water De-
partment, City of San Diego).

BS, Z; VF = 40,000 It (0.2 .sec) until Decemlier, 10.52; V =
40,000± (0.2 sec.) since February. 10.53.

Benioff capacity seismograph, EW and NS ; VF = 8000± (1
sec) since March, 19.53.

Woody; N = 35° .',2.0'; W= 118° .',0.0'; H = 500 (Kern County
Forestry and Fire Department), installed on August 5, 1952.
BS, Z; V = 30,000± (0.2 sec)

Fort Tejon; N = 3i° 52.1,'; W = 118° 53.7'; H = 980 (State
Board of Beaches and Parks, Fort Tejon Historical Monu-
ment, State of California), installed on November 21, 1952.
BS, Z; V =30,000± (0.2 sec)

The following are temporary installations ; all were
equipped with Benioff vertical seismographs and sliort-
period galvanometers; constants were average and re-
cording was on paper except for King Ranch, where re-
cording is on film :

Chuvhiipute; N = 31,° 1,8.5'; IV = lt9° 00.7'; H = 1.590 ; (Ranger
Station, I!. S. Forest Service), installed on .luly 21, 19,52, di.scoii-
tinued on Novemlier 19, 1952.

Ilaviltth; N=35° SO.O' ; ^¥=118° 31.0'; H = 990; (Ranger Sta-
tion, U. S. Forest Service), installed on .Inly 25, discontinued on
September 4, 19.52.

linos Ranch; N=S5° 29.0': ]V=I18° 31.7'; 11 = 1090; (Mr.
and Mrs. Charles Knox), installed on September 4, discontinued
on November 10, 19.52.

King Raneh ; N = 35° 19.7' ; W = 119° 1,1,.7' ; H = 670 ; (Elmer
King Ranch, Mr. Charles Willis), installed on October 16, 19.52.

Williams Ranch: N = 35° 17.9'; W=118° 36.7'; H = 430 ;
(Mr. and Mrs. Boyd Williams), installed on November 10, 1952,
discontinued on March 20, 19,53.

In addition, for short intervals, portable instruments
were set up in the epicentral area. Most of them con-
sisted of a Benioff Vertical variable reluctance seismo-
graph recording with a short-period galvanometer on
photographic paper (indicated by .1 in Table 1). How-
ever, at two instaUations (indicated by B) a Benioff
capacity horizontal seismograpli was used recording on
Sanborn heat sensitive paper by means of a hot stylus
recorder, and at one location (C) this type of recorder
was connected with a Benioff capacity vertical com-
ponent.

Tahle 1

. Installations with portahle instruments.

Location

North
Lat.

West
Long.

Eleva-
tion
meters

Instru-
ment
(see
text)

Period

(PDT)

1952

BED

35»05.7
34'>59.1
SS^IS.O
35°42,5
35°24.3
35°21.8

35''15.1
35^28.9
35°26.4

35°14.8
35"'09.3
34°59.6

n8°24.7
118°31.a
118°39.9
118°33.8
118'>29.0
118°22.9

118<'36.6
118''44.6
118'=43.8

118°36.5
119°28.2
119''11.0

1310
1570

620
2000

760
1150

820
430
910

825
720
435

A
A
A

A
A
A
B

A
A

C
A
B

July 21

White Oak

Wliite Wolf

Shirley Meadow

Walker Dump

Piute Ranch

Clear Creek

Kern Gorge

Parker Creek

Clear Creek Ranch

Elkhorn

San Emigdio

July 21-22
July 23-27
Aug. 13-14
Aug. 14, 19-20
Aug. 20-21;
Sept. 3-5
Aug. 21-22-
Aug. 27-28
Aug. 28-29;
Sept. 3-5
Sept. 3-5
Nov. 12-13
Nov. 12-13

NS;
37°

V =

20. i';

40±
W

(0 to 10 sec)
- 121° 38.0' ; H = 1282

B. Instruments of the Berkeley group of stations
(based mainly on information furnished by Mr. Charles
E. Herrick, Berkeley).

Berkeley; N = 37° 52.3': W = 122° 15.6': H = 81. Mailing ad-
dress : Seismological Station, tlniversity of California, Berke-
ley 4, California.

G, EW and NS ; V = 1.300 (6 sec.) ; Z ; V = 1000 (6 sec)

BS, Z; V = 30,000± (0.2 sec)

TS, EW and NS

Bosch-Omori, EW and

Mount Hamilton ; A' = .
(Lick Observatory).
BS, Z; V = 30,000d
TS, EW and NS
Palo Alto; N - 37° 25.1';
I'niversitv) .
BS, Z ; V = 20,000± (0.2 sec)
TS, EW and NS

San Francisco; N = 37° 1,6. i' ; W = 122° 27.2'; H = 100 (Uni-
versity of San Francisco).
TS, EW and NS

Ferndale; N = 40° 31,'; W = 121,° i6': H = 17 (City of Ferndale)
Bosch-Omori, EW and NS ; V= 40+ (0 to 10 sec)

(0.2 sec

W = 122° 10.8'; H = 83 (Stanford

156

Earthquakes in Kern County, 1952

[Bull. 171

OOct 16

CALTECH
SEISMOLOGICAL LABORATORY

0 permanent stotions
O semi -permanent
• portable units
1952

•Aug 5 —

July 25-Sept 4
Sept 4 -Nov 10 '

►Aug 13-14

Aug 27-28^

Aug 28-29^:^ Aug, 20-21

Sept 3-5 Aug 14 -_- sepi 3-5

Aug 19-20'^ ^•-'

Nov 10- Mar 20, 53-C

July 23-27^'€-A^? 21-22

FiuURK 2. Map showing stations in the epicentral area.

Fresno; N = 30° /,(!.!■; W = 1 19° J,7.8'; H = SS (Fi'esno State
(/ollege).
S; EW, NS and Z
Mineral: X = JiO' 21'; \V = 121" SS' ; 11 = 1405 (National Tark
Service ) .
US, Z; V = 4(),()00± (0.2 sec.)
TS, EW and NS
Areata: N = 1,0° 52.0'; W - 12^° 0.',.5' : H = 60 (Humboldt State
College).
S ; EW, NS and Z
Shasta; N = 40° .',1.T ; W = 122° 2S.3' : H = 312 (U. S. Coast
and Geodetic Surve.v in cooperation with the Biirean of
Reclamation) .
BS with galvanometer period of abont IJ sec; Z, F^W, NS ;
VP = 40,000± (1 sec.)

€. Other permanent statiuits

Santa Clara; N = 37° 21'; W = 121° 57'; H = 27 (University of
Santa Clara).
G; Z, EW, NS; V = 1000± (G± sec.)

Iliali: X - 3')° «S'
(ieodetic Survey
tnde 01).servatory ) .
ilc( 'omb-Homl)erg seism

H'= 123° 13': II
in cooperation witli

= 27 (T'. S. Coast and
the International Lati-

;i-;ipii ;

70 (O to

sec.)

Pio'Vire 1 shows the location of the permanent and
semipermanent seismolopieal stations in California cUir-
ino- 1952. The locations of the portable installations in
the epicentral area are shown in figure 2 in which also
the permanent and semipermanent stations are marked.

The writers of Part II-G, 7, 8, and 0 wish to acknowledge their
indebtedness to the Director of the United States Coast and
(Ieodetic Survey and of many individual stations, who have lent
their (U-iginal seismograms or sent copies, and provided valuable
data for interpretation ; particularly to the staff at Berkeley, for
a long series of seismograms of the University of California group
of stations, as well as for magnitudes of aftershocks determined
at Berkeley.

Our grateful thanks go to the organizations and individuals
who have provided facilities for and helped maintain the special
stations set up in Kern County, particularly to the United States
Forest Service, the Kern County Forestry and Fire Department,
the California State Division of Beaches and Parks, Mr. and
Mrs. Charles Knox, Mr. and Mrs. L. E. Williams and Mr. Charles
Willis (at King Ranch). Property owners and tenants who
courteously provided sites for the portable recorder were : Mr. and
Mrs. Glenn \. Hurst. Jlr. .Tim Rogers, Mr. and Mrs. Kerinit
Austin, Jlr. X. Berry, E. Hales Ranch, San pjinigdio Ranch. We
are also indebted to the officers of the Monolith I'ortland Cement
Company. Operators of the Pacific Telephone and Telegrapti Com-
pany were extremely helpful during the emergency in keeping
field parties in touch with Pasadena headquarters. Special service
and many courtesies were pnjvided by Mr. and Mrs. Orville House,
at Clear Creek Cafe. A radio receiver for use in the special
recording program was lent by the Gilman Scientific Instrument
Company of Pasadena.

All members of the Seismological Laboratory staff made sig-
nificant contributions to the extensive program of field recording,
measurement and interpretation of seismograms.

Chief responsibility for setting up and maintaining stations was
shared between Jlr. F. E. Lehner and Mr. Ralph Gilnnm ; the
latter had charge of this during the important month of August.

Mr. G. G. Shor contrilmted heavily to all parts of the program,
including working out preliminary epicenters.

The figures in Part II-6, 7, 8, and 9 have been drafted by Mr.
Gilman and Mr. J. M. Nordijuist.

6. EPICENTER AND ORIGIN TIME OF THE MAIN SHOCK ON JULY 21
AND TRAVEL TIMES OF MAJOR PHASES

JiY a. (JrTKN'BKRf;

TliP epicpnter of the main earthquake on July 21,
19')2. was determined a) from the arrival times of P
at near-by stations; b) from comparison of these times
with those found previously for shoclvs in the same
re^'ion in which, contrasting- with the present shock, the
motion was not so large that the light spot left the paper
shortly after the beginning and in which the onset of
the transverse waves (S) could be found on records of
several stations (mainly shocks no. 13-16, Gutenberg
1943, with origin times revised in 1951) ; c) from simi-
lar comparison with times in records of aftershocks
originating near the main shock, but for which seismo-
grams from portable or temporary stations at short
distances furnished additional data. For details of the
method see Gutenberg (1943, p. 502). If arrival times
of Pn (longitudinal wave leaving the source downward
and refracted twice at the Jlohorovieic discontinuity ;
see fig. 1) at the stations near the Sierra Nevada (Hai-
wee, Tinemaha, Eeno) are used, the effects of the dif-
ference in crustal structure at the station must be con-
sidered. For Pn, which has to go deeper down than
usual as a consequence of the Sierra Nevada root, this
may result in a delay of as much as 4 seconds.

Methods a) and b) give the following coordinates for
the epicenter :

Latitude 35°00' North; Longitude 119^02' West (1)

ilethod c) was applied by C. F. Riehter and confirmed
the result within about 1 minute of arc (or about 1 mile).
The origin time resulting from methods a) and b) is

July 21, 1952, ll'=52"'14.3^ GCT.

(2)

Values (1) and (2) are used in calculations. Another
way to calculate the origin time when most stations are
too distant to record the direct longitudinal wave }:>
(fig. 1) leaving the focus upward, but have a clear Pn
(distance A less than 6°), is to find the intercept time K
(extrapolated travel time at A ^ 0°) of the travel time
curve of Pn for each station in a number of aftershocks
near the main epicenter (within a fraction of a degree)
and assume that the focal depths and the values of K
are the same in all these shocks. The travel time i of Pn
is then given with very good approximation by

t = K+hA. (3)

From Dr. Riehter 's investigations of seven shocks
originating close to the main shock the following values
of K result with & = 1 :8.18 see /km : Riverside 5.4 ±: 0.1,
Big Bear 5.8 ± 0.2, Palomar 5.6 ±: 0.2, Dalton 5.2 ± 0.1,
China Lake 5.5 ± 0.3. Fresuo 4.8 ± 0.3, Ilaiwee 7.2 ±
0.2 see. For Boulder, Berkeley, Palo Alto and Lick, the
data were not sufficient to find separate values of K
and the average of 5.1 sec. (Gutenberg, 1951) was taken
for each. Origin times of the main earthquake are calcu-
lated on the assumption that the average velocity of the
direct longitudinal waves (p) is 6.34 km/sec and that
of Pn (refracted at the Mohorovicic discontinuity) is
8.18 km sec. The resulting individual times are listed
in table 1. Their average is

0 = 11:52:14.3 ±0.1 sec.

(4)

Finally, the method of least squares was applied to
the residuals on the assumption that the vahies of K
and the velocities of p aiul Pn are correct as given above.
The result is:

Latitude 35°00' ± f North ; Longitude 119°01' ±

li' West (5)

Origin time ll'>52'"14.2' ± 0.13* (6)

The systematic errojs wliich depend on the assumptions
are probably greater than the standard errors resulting
from the calculation. It should be kept in mind that our
knowledge of the velocity in the earth's crust is rather
incomplete, especially near the low-velocity layer to be
discussed later in this section, and that local effects of
tlie .sedimentary layers, batholiths, roots of mountains,
etc. accumulate to several seconds as indicated by the
differences in the value of K discussed above.

The depth of focus can not be found very accurately.
From other data for southern California shocks and
artificial explosions an approximate focal depth of 15
km was considered to fit best with an estimated uncer-
tainty of about rt 6 km.

For a detailed study of the original records, all sta-
tions with which the Pasadena station exchanges bulle-
tins were asked for their records of the main earthquake
and some records of the aftershocks. Records of 165

Table 1. Cnlcuhifcd origin times of mniii xhocj;, Juli/ 21,
seconds after ll':'>2'" GCT.

From p:

Santa Barbara
14.8

Pasadena

14.0

Mount Wilson
14.5

From Pn:

Haiwee
14.8

Palomar
13.7

China Lake
14.4

Lick
13.8

Riverside
14.4

Palo Alto
13. S

Fresno
14.2

Boulder
15.1

Big Bear
14.3

Berkeley
14.1

stations have been received. I'nfortuiiately, in quite a
number of instances no instrumental constants are avail-
able, and in some the direction of ground motion corre-
sponding to an upward motion of the trace on the record
is not known. However, nearly all records can be used
for the study of times, since at very few stations is the
time correction in doubt by more than ±1 second. At
many stations it is given to 0.1 second and was changing
by less than 1 second per day. Reports giving the time
of P or S are available from 40 additional stations,
either direct from the station or through the U. S. Coast
and Geodetic Survey or the International Central Office
at Strasbourg.

Arrival times and ground amplitudes of the major
phases have been determined as far as possible. Arrival
times of the main phases are given in table 2. Where
the magnification of instruments is not known, ratio of
amplitudes in aftershocks was determined. This made
it possible to determine the magnitude of smaller after-
shocks relative to one of the larger aftershocks — usuallv

(157)

158

Earthquakes in Kerk County, 1952

[Bull. 171

S;ll

s 1 ^

^ -;: a

"^a.

« o. g

'-C

o1^.

c
o

f, ts

^ 2 S

':j

—

«1^,

■^

-«J

-a

!:i

t,'

tsa.

■^

(Tl

0; .t;

■n

t4-l

I'art III
0

SEC

Seismology

50 60

100

-2
-3
-4

7
SEC

6
5
4
3
2

- July 21, 1952 <>

RESIDUALS FROM GUTENBERG

RICHTER

(1935) h = 25ikm

0--ll:52;l4.3 GCT

• o

°.

I O n ^

• *

« » » ^; APPROXIMATE EFFECT OP
>. '~^. l/" ^ 25km -ISkm

DEPTH " v' * *^

o • 9 ^JL *^ ** o

oo o

o \ o

O lO K \

oi •/^''°

o \ O^ ^^'"»

•O- o o o*oo

• • •

00/

o /^ o
y cm o

OOM • • X

# 0

o ox ^^ ■ ^^

^^ ■

X X

• •

o «

•

.. • •

o >

159

0
-I
-2
-3

-4

• OBSERVED TO NEAREST 0.1 SECOID o OBSERVED TO NEAREST SECOIO > FROM REPORTS

July 21,1952
0--ll:52-l43GCT

RESIDUALS FROM JEFFREYS-BULLEN (1940) h-33km

.^-

I >- o

-1

I- \

HALF TIME DIFFERENCE
BETIEEN h-}]KM AND h-OKM

DISTANCE, DEGREES-

10 20 30 40 50 60 70 80 90 100 110

FiGXjRE 2. Residuals of observed travel times of lonyitudiiial waves as function of epicentral distance.

that of July 23, 0'> or July 29, 7", for both of which
the magnitude is well known. Resulting' data are used
together with those found from amplitudes whieh are
ealeulated directly by use of the instrumental constants.

Records of the longitudinal waves in the main shock
written at distances of less than 20° confirm conclusions
drawn previously by the author concerning: the structure
of the upper 300 km of the earth's crust. Since discus-
sion of these phenomena requires more space than is
available here, a paper presenting data and results was
prepared for publication elsewhere (Gutenberg, 1954) ;
the following is a summary.

There are indications of at least two layers in the
earth's mantle exhibiting lower wave velocity than the
layers above or below, one at a depth between roughly 10
and at least 20 km, a second in the "ultrabasic material"
from a depth of about 30 or 40 km down to roughly 150
km for P and 250 km for /S, and possibly a third between
these two in a different material (gabbro?). It is tenta-

tively suggested that the low-velocity channels are the
effect of the increase in temperature with depth whieh,
in the depth ranges involved, surpasses tlie effect of the
pressure increase acting in the opposite direction (Gu-
tenberg, 1955). Laboratory experiments by Hughes and
Cross (1951) with various rock types indicate that in
most the rate of increase in velocity with depth is getting
smaller, and that at a depth of the order of 10 km the
velocity increase stops altogether. In a few samples, for
which the effect of still higher pressure and temperature
could be observed, a decrease in velocity was found.

The deepest low-velocity channel occurs in the depth
range in which the temperature approaches the melting
point of the rocks and which frequently has been con-
sidered to be the upper part of the asthenosphere. This
channel may well be the seat of viscous flow which has a
tendency to restore equilibrium in tectonically disturbed
areas with a time of relaxation of probably over a thou-
sand years.

160

Table 2.

Earthquakes in Kerx County, 1952

[Bull, 171

Ohsen-ed anival times (min.:sec.) of longHi'dinal and transverse waves for main earthqunlce July 21. 1952 Hour (11 or 12 GCT)
is omitted. An asterisk (*) indicates that times are taken from report of stations. A = dtstance w degrees.

Station

Santa Barbara

Pasadena

Mount Wilson

China Lake

Haiwee

Riverside

Fresno

Big Bear

Tineinaha

Palomar

Barrett

Mt. Hamilton

Santa Clara

Palo Alto

Boulder City

Berkeley

San Francisco

Reno

Ukiah

Mineral

Shasta

Ferndale

Tucson

Corvallis-

Butte

Bozeman

Chihuahua

Spokane

Seattle

Hungry Horse

Victoria

Lubbock

Alberni -

Lincoln

Saskatoon

Guadalajara

Fayetteville

Florissant

Saint Louis

Cape Girardeau

Tacubaya

Puebla

Sitka.-

Veracruz

Whiting Field

Cincinnati — -

Ann Arbor

Merida.

Cleveland

Columbia

Pittsburgh

Kirkland Lake

Buffalo-

Guatemala City

State College

Washington- -

Ottawa.

Miami

College

Philadelphia.

Hawaii

Swan Island

New York C.C

Palisades

Fordham.

Shawinigan Falls...

Honolulu

Harvard

Weston

Kingston

Resolute Bay

Guantanamo Bay,.

Halifax

Mitchel Field

Balboa..

Bermuda

San Juan

Roosevelt Roads. -

Bogota.-

Morne des Cadets-

Scoresbysund

Reykjavik

Huancayo

Apia-.

La Paz ' —

SKS

0.78

52:28.7

1.11

33.7

1.12

34.3

1.42

39.2

1.43

40.9

1.70

42.9

1.87

44.4

1.90

46.1

2.18

50.9

2.45

52.2

3.04

53:00.1?

3.15

01.7

3.33

3.50

OR. 2

3.59

09.0

3.88

11.9

3.89

12.9

4.56

24.8

5.28

5.72

40 ±

6.30

45.9

6.93

58 ±

7.35

54:01.0

10.10

43.2

12.05

55:11.1

12.29

12.70

55:18±

12.78

12.88

13.84

33.2

13.89

32.6

14.28

14.89

18.61

56:35

19.30

19.85

20.22

50.2

23.17

57:23*

23.32

24*

23.9

30*

23.45

24.40

25.07

25.7

27.3

58:01.6

27.7

05*

28.4

29.4

24 ±

30.0

31.2

31.7

32.0

43*

32.8

52*

32.8

33.5

57*

34.3

59:01.5

34.4

05.7

34.5

59:03

34.9

07*

35.5

14.6

35.6

15.0

35.7

15*

35.7

35.8

11*

36.4

36.6

23.7

37.4

28*

37.5

31.1

41.1

00:01

41.5

03.8

42.9

15M

43.9

22*

44.4

44.8

30.4*

49.5

01:05

50.0

11.1

51.3

21*

55.5

50*

59.7

02:18*

62.1

62.5

40H

69 . 7

03:25

70.4

31.5

Station

52 : 50 ±

00:14
30

01:36*
39*
49*
44

02:03±
02

19?

39*

53*

:11

:12

15±

07:11*
08:17

09:30
11:07

12:34
34*

Angra do Heroismo__

Kiruna

Sapporo

Aberdeen

Edinburgh. —

Rathf arnhani

Bergen

Mizusawa

Durham

Sendai

Tokyo CMO

Tokvo ERI

Kew

Uppsala

Matsushiro

Jersey

W'ajiina

Nagoya

De Bilt

Witteveen

Copenhagen

Hamburg

Kaniigamo

Coimbra

Paris

Osaka

Santa Lucia

Kobe

Lisbon

Koti

Jena

Clermont-Ferrand _ . .

CoUmberg

Strasbourg

Toledo

Stuttgart

Basel

Cheb

Neuchatel

Zurich

Praha

Fukuoka

Miyazaki

Tortosa

Chur

Malaga

Cartuja

Averroes

Barcelona

Milano

Pavia

Ahneria

Alicante

Sal6

Wien

Skalnate Pleso

Bologna

Trieste

Hurbanovo

Padova

Prato

Firenze-Ximeniano _

Budapest

Eva Per6n

Zagreb

Kalocsa

Alger-University

M'Bour- -.

Roma

Rocca di Papa

Beograd

Napoli

Tunis.- -

Taranto

Messina

Wellington

Istanbul

Athens

Christchurch

Tamanr asset

Brisbane

Baguio

Hongkong

Manila

SKS

70.4

73.1

73.3

74.0

74.3

74.5

74.6

75.5

75.8

76.2

78.3

78.5

78.9

79.2

79.3

80.5

80.6

81.4

81.6

81.7
81.8
81.8
82.0
82.1
83.7
84.0
84.1
84.3
84.3
84.4
84.7
85.0
85.0
85.1
85.5
85.8
85.8
86.2
86.3
86.3
86.3
86.4
86.6
86.7
86.9
87.4
87.4
87.5
87.7
88.1
88.7
89.0
89.1
89.1
89.2
89.3
89.5
89.7
89.9
89.9
90.5
90.6
91.4
91.5
91.7
92.5
93.2
94.4
94.8
95.9
97.2
98.8
99.6
99.9
102.1
103.7
103.8
103.9
104.7

30
44.9
03:47Ji
50
53

54.5
55
04:00
02*
03.5
17.0
17}i
17 >^
19

21d=
21*
24*
27

29.3
30*
29.7
32
33*
34

34.8
36.6
34?
39?
37
45*
04:46
48*
48.0
48.2
48
50.3
51.21
52*
51.8
54.4
56.2
55.5
59
59

58.7
58*
05:01

03.6
04:59
05:00±
04.9
00*
03*
03.3*
07
11*
10±
10.9
13*
05*
05:09H
10*
14H
13

14.8
18*
17.6

24*

26.0

34*

40.3'

56.5

06:00
11.1
18±
17
23*
20

15:10
16*
(12)
(15)

18
21
28
26
29

22?

33*

25*

42*
47
40
41*
47*
15:40

45
43

52 ±

45*

16:01±

35 ±
13:12
13:15

46*

48
14:18

25*

26*

57 H*

52
15:10
15:12

26*

23*

31?

34?

38?

48?

43?

49*

55
16:10*
15:55

56*

16:01
08*
02?
13
06
26*
20

17±

11*

17"
18.4*
22
30
34
27
44
51
57
17:00*

17:01*
15

18:02?
07

Part III

Seismology

101

Table 2. Ohscrrctt urririil liiiien fniui.:xec.) of iomjitmUnul and

tninsieixe iriiifx for niiiiii eiirlhiiunke ■hilij 21, 1U.',2. Hour (II or

12 flCl'J is oiiiiltril. Aii asterisk (*) iiiilicntes Unit times are t<ik<ii

from report of stations. A ~ distance in degrees. — Continued.

Station

Ksara

Rivervievv, .

Helwan

Chatra

Calcutta.

Hyderabad

Poona

Bandung

Djakarta

Hermanus

Kiniberley

Pretoria

Grahamstown

Pietermaritzburg
Tananarive

107.6
108.3
109.7
113.7
117.2
125.2
125.4
128.8
129.0
145.9
148.8
150.3
151.5
153.6
160.1

06:37
43

10:29

54
57

11:17.3

23?

58.6
12:01

04?

18±

SKS

17:16
17

17:51

18:32?

18:34

•

1 1 — 1 ■ — r 1 ,

, S RESIDUALS • • -
FROM

lU
8

„ 0

JEFFREYS-BULLEN (1940) • • ° •• -

o o

« OOo •
•o-

' i ...9 \i' 2i° ;

-2

— •

• O

• * -

o
A —

• from report •
f secorjd phose

1 1 1 I 1

20'

30*

40"

50'

60' 70°

80°

90"

Figure 3. Residuals in seconds of observed travel times of
transverse waves in the main shock of July 21, l!iri2. as function of
ei)icentral distance. The residuals are based on the mean of the
travel times of Jeffreys-Bullen (1940) for surface focus and for
h = 0.00 (depth 33 km.).

"Waves originating in a loAV-velocity channel are pre-
vented from leaving this channel if they start out in a
direction not too far from the direction of its axis. Paths
of waves entering a low velocity layer from outside di-
verge inside such a channel, so that the energy flowing
through a square centimeter of the wave front decreases
rapidly with distance. If an earthcjuake originates in a
low- velocity layer (as indicated in fig. 1), waves leaving
the source downward cannot enter the layers with rela-
tively high velocity above the channel and consequently
cannot reach the surface of the earth unless they have
reached a layer below the source with a wave velocity at
least equal to the maximum velocity above the source.
Thus a large fraction of the energy remains in the chan-
nel. In earthquakes with a focal depth of about 12 km
or less, that is, in shocks originating above the upper
channel, much more energy is transmitted to the cpi-
central area than in those originating at greater depth.
This may contribute to the unexpectedlj' great damage
occurring sometimes in relatively small but rather shal-
low shocks.

Arrival times of longitudinal and transverse waves
from the main shock on July 21 and of the largest after-
shock on July 29 recorded at epicentral distances of less
than 18° (less than about 1.250 miles) are found to line
up along complicated travel time curves as a consequence

of tlie pec'uliar ray patlis produced by the low velocity
layers. At epicentral distances between about 5° and 15°
the first waves of the main shock arrive relatively late
and show very small amplitudes; starting at about 15°
or 16° the amplitudes of the first P-waves are larger
than at a distance of 6°. S-waves decrease similarly at
short distances; no S-waves have been found between
about 10° and 18°. but at a distance of 19° they have
relatively large amplitudes. The difference in distance
at which the two types of waves reappear with large
amplitudes is a conse(juence of the noticeable increase in
Poisson's ratio in and below the low velocity channel at
depths between about 100 and 300 km. This indicates
that in this depth range the bulk modulus (resistance to
compression) of the rocks increases faster (or decreases
less) than their rigidity (resistance to shearing).

Residuals of observed P-waves relative to the travel
times of Gutenberg-Richter (1934) and to those of Jef-
freys-Bullen (1940) are reproduced in figure 2. The
former are calculated for a focal depth of 25 km. the
latter for 33 km. The effect of the difference in depth
between the values in the tables and those for the depth
of 15 km (considered to be the best approximation for
the main shock) is indicated in the figure. The residuals
indicated in figure 2 have been combined with similar
residuals observed in longitudinal waves from Pacific
surface foci (Gutenberg, 1953). The resulting correc-
tions have been applied to travel times for continental
surface foci by Gutenberg and Richter (1939, Table 1,
p. 97). By subtracting from these travel times the time
difference corresponding to a difference in depth between
zero and 25 km. new tables for travel times of P for a
focal depth of 25 km have been calculated by ^Mr. John
Xordquist, and a new table was set up to give epicentral

30
SEC

July 21, 1952 ^ r-
>

i A .

. r^

/ ' /°

—

I / /

- CO

y y y

- <

y ' ix

■ o

" <M
CJ

^ °io / CP^

V °

/ 0^ 0 °

• 0

/ '*o

• •/

O \

•/

SKsN-g° ■

/•

. „., ■ \ :

o e y

o from reporl \

-in

f second phose V

A — »

0" 82"

84* 86" SB' 90* 92

Figure 4. Ob.served travel time of H and SA'.S' in the main shock
of July 21, 1952, minus 22 minutes and 20 seconds and minus Ti
times distance in dejirees beyond 80 degrees as function of distance.
The fi^'ure shows the interscctiou of the travel time curve for S
with that for HKS.

162

Earthquakes in Kern County, 1952

Table S. Epicentral distances for given travel times of P and focal depth of 25 km. 1953 revision.

[Bull. 171

P-O

Diff. in d

(min.

for

sec.)

Ah=+10km.

0.00

0.0

0.1

0.2

0.3

0.4

_ ,

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

— .05

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

+ .02

1.9

2.0

2.1

2.2

2.3

2.4

2,5

+ .04

2.6

2.7

2.8

2.9

3.0

3.1

3.2

+ .05

3.3

3.4

3.5

3.6

3.7

3.8

3.9

4.0

+ .06

1.00

4.1

4.2

4.3

4.4

4.5

4.6

4.7

+ .06

4.8

4.9

5.0

5.1

5.2

5.3

5.4

5,4

.06

5.5

5.6

5.7

5.8

5.9

6.0

6.1

6,1

.06

6.2

6.3

6.4

6.5

0.5

6.6

6.7

6.8

.06

6.9

7.0

7.1

7.2

7.3

7.4

7.5

.06

7.5

7.6

7.7

7.8

7.9

8.0

8.1

.06

2.00

8.2

8.3

8.4

8.5

8.6

8.7

8.8

+ .06

8.9

9.0

9.1

9.2

9.3

9.4

.07

9.5

9.6

9.7

9.8

9.9

9,9

10.0

10.1

.07

10.1

10.2

10.3

10.4

10.5

10.6

10.7

.07

10.8

10.9

11.0

11.1

11.2

11.3

11.4

11.5

.07

11.5

11.0

11.7

11.8

11.9

12.0

12.1

12.2

.07

3.00

12.3

12.4

12.5

12.6

12.7

12.8

12.9

13.0

+ .07

13.1

13.2

13.3

13.4

13.5

13.6

13.7

13.8

.07

13.8

13.9

14.0

14.1

14.2

14.3

14.4

14.5

14.6

.07

14.7

14.8

14.9

15.0

15,1

15.2

15.3

15.4

.07

15.5

15.6

15.7

15.8

15.9

16.0

16.1

16.2

.08

16.2

lfi.3

16.4

16.5

16.6

10.7

16.7

16.8

16.9

17.0

.08

4.00

17.1

17.2

17.3

17.4

17.5

17.6

17.7

17.8

17.9

+ .08

17.9

18.0

18.1

18.2

18.3

18.4

18.5

18,6

18.7

.08

18.8

18.9

19.0

19.1

19.2

19.3

19.4

19.5

19.6

.08

19.7

19.8

19.9

20.0

20.1

20.2

20.3

20.4

20.5

.09

20.5

20.6

20.7

20.8

20.9

21,0

21,1

21.2

21.3

21.4

.09

21.5

21.6

21.7

21.8

21.9

22.0

22.1

22.2

22.3

22.4

.09

5.00

22.5

22.6

22.7

22.8

22.9

23.0

23.1

23.2

23.3

23.4

+ .09

23.5

23.6

23.7

23.8

23.9

24.0

24.1

24.2

24.3

.09

24.4

24.5

24.6

24.7

24.8

24.9

25.0

25.1

25.2

25.3

.10

25.4

25.6

25.7

25.8

25.9

26.0

20.1

26.2

26.3

26.4

.10

26.6

26.7

26.8

26,9

27.0

27.1

27.2

27.3

27.4

27,6

.10

27.7

27.8

27.9

28.0

28.1

28.2

28.3

28.4

28.6

28.7

.10

6.00

28.8

28.9

29.0

29.1

29.2

29.3

29.4

29.6

29.7

29.8

+ .10

29.9

30.0

30.1

30.2

30.3

30.4

30.6

30.7

30.8

30.9

.10

31.0

31.1

31.2

31.3

31.4

31.0

31.7

31.8

31.9

32.0

.10

32.1

32.2

32,3

32.4

32.0

32.7

32.8

32.9

33.0

33.1

.10

33.2

33.4

33.5

33.6

33.8

33.9

34.0

34.1

34.2

34.3

.11

34.4

34.6

34.7

34.8

34.9

35.0

35.1

35.2

35.4

35.5

.11

7.00

35.6

35.8

35.9

36.0

30.1

30.2

36.4

36.5

36.6

36,8

+ .12

36.9

37.0

37.1

37.2

37.4

37.5

37.6

37.8

37.9

38,0

.12

38.1

38.2

38.3

38.4

38,0

38.7

38.8

38.9

39.0

39,1

.12

39.2

39.4

39.5

39.6

39.8

39.9

40.0

40,1

40.2

40 4

.12

40.5

40.6

40.8

40.9

41.0

41.1

41.2

41.4

41.5

41.6

.13

41.8

41.9

42.0

42.1

42.2

42.3

42.4

42.6

42.7

42.8

.13

8.00

42.9

43.0

43.1

43.2

43.4

43.5

43.6

43.8

43.9

44.0

+ .14

44.1

44.2

44.4

44.5

44.6

44.8

44.9

45,0

45.1

45.2

.14

45.4

45.5

45.6

45.8

45.9

46.0

46.1

46.3

46.4

46.0

.15

46.7

46.9

47.0

47.1

47.2

47.4

47,5

47.0

47,8

47.9

.15

48.0

48.1

48.3

48.4

48.6

48,7

48,9

49.0

49,1

49.2

.15

49.4

49.5

49.6

49.7

49.9

50.0

50.1

50.3

50.4

50.0

.16

9.00

50.7

50.9

51.0

51.1

51.2

51.4

51.5

51.0

51.8

51.9

+ .16

52.0

52.1

52.3

52.4

52.6

52.7

52.9

53.0

53.1

53.3

.17

53.4

53.6

53.7

53.9

54,0

54.1

54.3

54.4

54.6

54,7

.17

54.9

55.0

55.1

55.3

55.4

55.6

55.7

55.9

56.0

50,1

.18

56.3

56.4

56.6

56.7

56.9

57.0

57.1

57.3

57.4

57,6

.18

57.7

57.9

58.0

58.1

58.3

58.4

.i8.6

58.7

58.9

59,0

.18

10.00

59.1

59.3

59.4

59.6

59.7

59.9

00.0

60.1

60.3

60.4

+ .18

60.6

60.7

60.9

61.0

61.1

61.3

61.4

61.6

61.7

61.9

.18

62.0

62.1

62.3

02.4

62,6

62.7

62.9

63.0

63.1

63.3

.18

63.4

63.6

63.7

03.9

64.0

64.2

64.3

64.5

64.7

64.8

.18

65.0

65.1

65.3

05.4

65.6

05.7

65.9

66.0

66.2

66.3

.18

66.5

66.7

66.8

67.0

67.1

07.3

67.4

67.6

67.7

67.9

.18

11.00

68.0

68.2

68.3

68,5

68.7

08.8

69.0

69.2

69.3

69.5

+ .18

69.7

69.8

70.0

70.2

70,3

70,5

70.7

70.8

71.0

71.2

.18

71.3

71.5

71.7

71.9

72.0

72.2

72.3

72.5

72.7

72.8

.19

73.0

73.2

73.3

73.5

73.7

73.8

74.0

74.2

74.3

74.5

.19

74.7

74.8

75.0

75.2

75.3

75.5

75,7

75.8

76.0

76.2

.20

76.3

76.5

70.7

76.8

77.0

77.2

77.3

77.5

77.7

77.8

.20

Part 11] Seismology 163

Table 3. EpicenfrnI distarices for given travel times of P and focal depth of ^5 km. 1953 revision. — Continued.

P-0

DifT. in A

(min.

for

sec.)

Ah=+10 km.

12.00

78.0

78.2

78.4

78.6

78.8

79.0

79.2

79.3

79.5

79 7

+ .20

79.8

80.0

80.2

80.4

80,6

80.8

81.0

81.2

81.3

81.5

.20

81.7

81.8

82.0

82.2

82.4

82.6

82.8

83 0

83.2

83.4

.21

83.6

83.8

84.0

84.2

84.4

84.6

84.8

85,0

85.2

85.4

.22

85.6

85.8

86.0

86.2

86.4

86.6

86.8

87.0

87.2

87 5

.23

87.8

88.0

88.2

88.4

88.6

88.8

89.0

89.2

89.5

89.8

.24

13.00

90.0

90.2

90.4

90.6

90.8

91.0

91.2

91.4

91.6

91.8

+ .25

92.0

92.2

92.5

92.8

93.0

93.2

93.4

93.6

93.8

94.0

.25

94.2

94.5

94.8

95.0

95.2

95.4

95.6

95.8

96.0

96.2

.26

96.5

96.8

97.0

97.2

97 4

97.6

97.8

98.0

98.2

98.5

.26

98.8

99.0

99.2

99.4

99.6

99.8

100.0

100.2

100.5

100.8

.27

101.0

101.2

101.4

101.6

101.8

102.0

102.2

102.5

102.8

103.0

.28

14.00

103.2

103.4

103.6

103.8

104.0

104.2

104.5

104.8

105.0

105.2

+ .28

105.4

105.6

105.8

106.0

106.2

106.5

106.8

107 0

107 2

107.4

.28

107.6

107.8

108.0

108.2

108.5

108.8

109 0

109,2

109.5

109.7

.28

109.9

110.1

110.3

110.6

110.8

111.0

111.2

111,4

111.7

111.9

.28

112.1

112.3

112.6

112.8

113.1

113.3

113.5

113.7

114.0

114.2

.29

114.4

114.6

114.8

115.1

115.3

115.5

115.7

US. 9

116.2

116.4

.30

15.00

116.6

116.8

117.0

117.3

117.5

117.7

117.9

118.1

118.4

118.6

.30

distances correspouding to a given travel time of P in a
continental shock with a focal depth of 25 km. These
values which are needed to locate earthquakes, if travel
times of P from various stations are known for a given
shallow shock of average depth, are reproduced in
table 3.

Residuals of the observed travel times of S-waves
from the travel time curves of Jeffreys-Bullen (1940)
are shown in figure 3. It is assumed that averages be-
tween the travel times for surface focus and those given
for a depth of 33 km — marked by 0.00 in the tables — •
correspond to the depth of focus of the main shock. The
resulting residuals are relatively small. At most dis-
tances the first S-waves seem to arrive 1 to 2 seconds
later than given by the tables, but on account of the
po.ssible error in the focal depth and of the scattering
of the residuals no attempt has been made to improve
the existing tables.

Of special interest are the observed times of the
phases S and SKS at epicentral distances near 84°
where the travel time curves of these two phases inter-
sect. SKS waves are transverse in the mantle, longi-
tudinal in the earth's core (see figure 7, Part II-l, in
General introduction to seismology). At distances
greater than about 84°, the SKS waves precede the
S-waves which have travelled over less curved paths
through the mantle only. In figure 4 the observed travel
times of both phases are plotted after deduction of the
quantity 22°>"' 20^"'^ — 7.5 (A-80) to permit use of a
larger scale. A is here the epicentral distance of the
station in degrees. Travel times found by Nelson (1953)
for SKS agree with those indicated by the curve in fig-
ure 4 within ±2 seconds.

Observed travel times for P', the longitudinal wave
through mantle and core, are given in table 4. The re-
siduals refer to the travel times given by Gutenberg and
Richter (1939, Table 19, p. 115) for surface focus di-
minished by 4 seconds for focal depth. The residuals are
within the limits of error.

The phase SKP (first path through mantle transverse,
path through core and second through mantle longi-
tudinal) is recorded with large amplitudes at Kodai-

kanal (A=132.4°) near its focal point. At Perth (A=
134.6°) the amplitudes are still rather large. The ob-
served travel times of 22""" :4P''^^ and 22:56 respectively
agree with those calculated from the tables within the
accuracy with which Milne-Shaw records ( 1 mm on the
trace corresponds to 7 J sec) can be read. Travel times
of "channel waves" are discussed in a special study by
Gutenberg (1955). No travel times of body waves other
than those mentioned or of surface waves have been
investigated thus far. The records of the earthquake are
an extremely valuable source of research material for
the future. However, periods of the longest surface
waves (G) have been measured. At a few stations the
longest recorded G-waves had periods of about 1 minute,
although maximum periods of about f minute were ob-
served more frequently.

Amplitudes of P, PP, S and of surface waves have
been used to determine the magnitude M of the shock.
Results are given in table 2 of Part 1 1-8 Magnitude de-
termination for larger Kern County sJiocks and are dis-
cussed there. The deviations from the average do not
exceed the limits of error by amounts large enough to
indicate significant corrections to the tables and graphs
which are based on the amplitudes of body waves at
different distances from a given source and are used to
determine the magnitude of a shock on the basis of ob-
served amplitudes and periods of P, PP and ;S at a given
distance.

Table -i. Observed residuals of P'.

Station

Helwan

Chatra.._

Poona

Bandung

Djakarta

Hermanns

Kimberley

Pretoria

Grahamstown

Pietermaritzburg
Tananarive

A
degrees

109.7
113.7
125.4
128.8
129.0
145.9
148.8
150.3
151.5
153.6
160.1

Travel

time
min:sec

18:40

18:43

19:03

19:13

19:09?

19:42

19:45

19:47

19:50

19:50?

20;04±

Residuals
sec

+ 8
+ 2
—2
+ 1
—3?
+ 2

+ 2
+ 2
—2?
+ 3±

7. THE FIRST MOTION IN LONGITUDINAL AND TRANSVERSE WAVES OF THE
MAIN SHOCK AND THE DIRECTION OF SLIP

HV B. (UlTEN'UKKO

ABSTRACT
Data on compressions and <lilnt!iti<ins in tlic ilircot longitudinal
waves are given and used for tin' detcrniinalion of the fault |)lnne
at the starting point of the eiirth(|uake and for the tindiiiK of the
direction of slip. A method is deveh)ped to ^et similar information
from the first motion in transverse waves reeorded at stations in
the hemisphere around the epieenter ami is applied to observed
amplitudes which are listed. The final results are: at the depth of
the source (about 10 miles) the fault plane has a dip of about
60° to 66° towards E 50° S ; the slip aloiiK the fault at this depth
was roughly up towards north in the upper ( southeastern ) block
relative to the lower (northwestern) block; the vertical com-
ponent of the slip was about 1.4 times that of the horizontal ; the
horizontal component corresponds to a relative movement north-
eastward in the upper block (southeast of the fault), southwe.st-
ward in the lower block.

Important information eoncerning the mechanism of
faulting: ^'aii frequently be obtained from studies of the
direction of first motion in longitudinal (P) and trans-
verse (S) waves at a sufficient number of well distrib-
uted stations.

For a study of compressions and dilatations as indi-
cated by the first onset of longitudinal waves in earth-
quake records the direct p (epicentral distances less
than about 140 km) can be used, the wave Pn (see fig. 1,
Part II-6,) where it is clearly recorded, that is, at dis-
tances not over about 600 km, the wave P at epicentral
distances between about 16° and 100°, and P' (through
the core). In the main shock of July 21 p and Pn started
with a dilatation at all stations, except perhaps for
Riverside, where the first very short motion is small and
possibly a compression. At the Big Bear station, about
50 km northeast of Riverside, the beginning is small,
but a clear dilatation. At epicentral distances of about
600 to 1600 km there is a shadow zone for longitudinal
waves (see fig. 1, Part II-6). It is not known how the
first waves arriving in this zone have traveled; if they
have been reflected somewhere, compressions may have
been changed to dilatations, and vice versa. For this
reason, waves arriving in the shadow zone around the
epicenter are not used for determination of compres-
sions and dilatations even in the rare instances where
the first wave in the seismogram is large enough to per-
mit the finding of the direction beyond reasonable doubt.

Contrasting with the dilatations at the near-by sta-
tions, the onset of P at 62 stations beyond the shadow
zone corresponds to a compression and at 6 additional
stations probably to a compression. Scattered among
these compressions are dilatations at Tortosa and at
Tamanrasset and a doubtful beginning at Cartuja
which reports a dilatation. All records beginning with
a clear P' (longitudinal wave through the earth's core)
indicate compression.

Clear dilatations in the first P-wave were reeorded at
"Whiting Field, Swan Island and Miami ; these stations
are under the supervision of the U. S. Fleet Weather
Central at Miami and have been equipped with very
sensitive instruments by the U. S. Navy Department,
for the investigation of microseisms in the Caribbean
area. Tlie records written at Guantanamo Bay and
Roosevelt Roads by similar instruments begin with a
clear compression. At the U. S. Coast and Geodetic

Survey station at San Juan the first motion is probably
a small comiiression followed by a large dilatation;
however, the first half wave is scarcely larger than the
biickoround of microseisms. The first longitudinal waves
on the records of the Mexican stations are rather small,
but all seem to correspond to dilatations. At Saskatoon
the records seem to begin with a small dilatation, fol-
lowed by a large compression.

Data for compressions and dilatations in the after-
shocks are much more scanty and, with few exceptions,
are limited to near-bv stations. They will be discu.ssed
by C. F. Richter in Part II-9.

A given motion at the source produces a unitiiie pat-
tern of compressions and dilatations at the surface of the
earth. Our problem is to deduce the direction of this mo-
tion from the observed pattern of compressions and dila-
tations. There are two difficulties involved in this task.
One is that the observations are limited to certain spots
scattered over the surface of the earth and separated
by large areas of oceans, by regions which have no
stations or do not give otit information, and by "shadow
zones. ' ' The other is a consequence of the fact that
relatively simple assumptions have to be made to make
a theoretical treatment possible. For example, it is gen-
erally assumed that the fault is a plane. Actually, in
many instances there is good evidence that the dip of
the fault surface changes with depth ; if it changes along
the fault, the Intersection of the fault surface with the
horizontal plane through the focus is not parallel to the
surface trace. In any case, direction of motion calculated
from the pattern of compressions and dilatations cor-
responds approximately to the direction of motion at
the point at depth where fracturing has started.

If we assume dip slip motion along a plane fault
having a dip angle 8 (fig. 1), we should observe at
the earth's .surface two sectors with compressions and
two with dilatations. These four sectors are then sep-
arated by the fault plane and an auxiliary plane per-
pendicular to it through the earthquake focus as indi-
cated In figure 1. The width of the zone near the
epicenter E exhibiting compressions in figure 1 is given
by 2/i/sin 28, if h is the focal depth, ancl the curvature
of the earth can be neglected. If h is small, this zone is
usually rather narrow. However, the curvature of the

E =EPICENTER

7«

COMPRESSION

Figure 1. Sketch of distribution of compressions and
dilatations in an earthquake.

(165)

166

Earthquakes in Kern County, l!)r)2

fr.uU. 171

'^•^

JULY 21. 1952

STATIONS PLOTTED AT

EXTENDED DISTANCES

( Hodgson, 1953)

r- I 00

GOOD D0UB1
COMPRESSION •

DILATATION T

Figure 2. Observed compressions and dilatations in the main earthquake of July 21, 1952. For the projection,
see text. The shadow zone for longitudinal waves is indicated by shadinR.

rays ha.s always to be considered. Nearly all rays of
seismic waves emanating: from the source intersect the
earth 's surface at shorter distances than the straight
lines tangent to them at the source. Thus, the ray
leaving the focus downward along the fault surface in
figure 1 and forming the boundary between compres-
sions and dilatations arrives at the surface at a point Si
much closer to the epicenter that the point S/ on the
straight line extending the fault plane. In order to find
in which quadrant of dilatations or compressions at the
source a raj' starts which arrives at a .station Ss (fig. 1)
one cannot use the location of .S'a relative to the straight
lines (planes) in the figure, but must find the direction
at which the ray leaves the source. This is given by the
tangent to the ray at the source which, in the figure,
intersects the surface of the earth at the point S3' ; this
point has been called "extended position" of S3 by
Byerly (1922). Tables for the "extended distances"
(e.g. arc S3' — E) of these extended positions from the
epicenter, if the actual distances (e.g. S3 — E) are
given, have been calculated by Hodgson and Storey

(1953). For the reasons given above, distances between
6° and 16° should not be used ; Hodgson and Storey
have already realized that difficulties arise for small
distances. For distances less than 6° the vertical distri-
bution of wave velocities near the source and the depth
of focus produce considerable differences.

Use of stereographic projection simplifies the study
of compressions antl dilatations. The tables of Hodgson-
Storey include the transformation of distances along the
earth's surface to those in a stereographic projection;
the unit of length used in these tables is the radius of
the earth. Hodgson and JMilne (1951) have summarized
earlier work, especially results of Byerly and of Adkins,
and have improved the method. For details of the theory
and its application to observations, the references should
be consulted.

In figure 2 compressions and dilatations are plotted
for the main sliock of July 21. The location of the points
is given by the station azimuths, taken at the epicenter
and the extended distances of the stations. For conver-
sion of the epiceutral distances into extended distances,

Part 11

Seismology

167

the llodfrson-Storey tables are used. It sliould be kept
in luinti that these tables are based on certain assump-
tions coneerninw the velocities of lono;itudinal waves in
the earth's interior (which are considered to be good
approximations for most distances) and that figure 2
is a stereographic projection of the "extended dis-
tances" and thus distorted in a way depending on the
change of longitudinal velocity with depth. The inter-
sections of the fault plane and of the auxiliary plane
with the earth's surface remain circles in the projection.

Frequently, nothing is known about the dip and
direction of motion in an earthquake to be investigated,
and the circles separating compressions and dilatations
on the projection may be drawn in a variety of ways.
Fortunately, in our case it is known from geological
investigations that the strike of the White Wolf fault
is approximately towards N. 50° E. In addition, the
main shock and all larger aftershocks during the first
36 hours which have been located are southeast of the
fault trace at a distance from it which is smaller than
the focal depth. Consequently, it can be assumed that
the fault dips rather .steeply, approximately towards
southeast, and that in the projection the center of the
circle corresponding to the intersection of the faidt plane
with the earth's surface is about southeast of the epi-
center. In the projection this circle should be tangent
to the fault trace. There is little choice to draw a circle
which fulfills these requirements and, in addition, in-
cludes onh^ dilatations. The fault plane circle indicated
in figure 2, consequently, can be assumed to be with
good approximation the projection of the intersection
of the fault plane with the earth's surface. Since its
diameter d in the units used by Hodgson-Storey is about
1.9 and tan 8 = d, it follows that the dip angle 8 of the
fault plane with a horizontal plane is about 63° (with
an estimated error of less than ±5°). This woidd not
disagree with the geological evidence and would cor-
respond to the estimate based on the location of epi-
centers relative to the fault trace. The resulting rela-
tive motion is downward in the lower block, upward in
the upper block, as indicated in figure 1.

In case of dip-slip, the maximum width of the zone
with compressions surrounding the epicenter was found
above to be given by 2/i/sin 28. With h — 15 km and
6 = 63° this gives about 37 km. There was no station
at so short a distance from the epicenter. However, the
result would be different if the motion had a component
in the direction of the strike. This can be found theo-
retically from the second circle which separates com-
pressions and dilatations. This auxiliary circle is the
intersection of the auxiliary plane (figure 1) with the
earth's surface. Our data for constructing the projection
of this circle in figure 2 are less complete than those
for the projection of the fault plane circle, partly as a
consequence of the shadow zone which is marked in the
figure, partly due to the lack of not too distant stations
(except for Honolulu) in the southwestern half of the
map. In case of dip-slip the fault plane and the auxiliary
plane are perpendicular to each other, and the dip of the
auxiliary plane and the fault plane dip must add up to
90°. In the projection, the centers of the two circles and
the epicenter are then on one line, and the diameter of
the auxiliary circle in the units used is given by cot

Figure 3. Sketch of relationship between
the direction of P. S, SV, and SB. The azimuth
of the plane of propagation is indicated by 7,
the polarisation angle of S by t, the angle of
incidence of the ray by i.

63° = 0.51. The auxiliary circle marked a) in figure 2
fulfills these requirements. It includes all dilatations
established beyond reasonable doubt northwest of the
fault trace and no compressions and, therefore, repre-
sents a po.ssible solution.

If the direction of slip has a component in the direc-
tion of the fault strike, the center of the auxiliary circle
is not on a line perpendicular to the direction of the
strike, but, in the projection, the auxiliary circle still
must pass through the two points indicated in figure 2.
The circles b), c) and d) in figure 2 with the centers
B, C and D respectively represent possible solutions. If
the Saskatoon record starts with a small dilatation, a
circle slightly larger than d would be most likely. In this
case the motion along the fault surface at the Source
woiild have been almost south-north in the upper block,
that is, it would have had a strike component north-
eastward. However, the data for compressions and dila-
tations in P do not permit finding the orientation of
the auxiliary circle relative to the fault more accurately.
For the location of the fault circle, the recorded S-waves
give additional information.

The use of transverse waves (S) in finding the direc-
tion of motion in slip is more complicated than that of
longitudinal waves (P). The vibrations for P are theo-
retically in the direction of the ray which can be estab-
lished theoretically with fair approximation, if focus
and station are given, whereas those of S may be in any
direction perpendicular to the ray. For calculations con-
cerning amplitudes of ;8, the motion in S is usually
separated into two components which are respectively
in the plane of propagation {SV) and perpendicular
to it (SH) (fig. 3). The motion in SV is perpendicu-
lar to the motion of P. SV has a horizontal component
in the same azimuth as P and a vertical component ; one
of the two has a direction opposite to the corresponding
component in P. SH has only a horizontal component
in a direction perpendicular to that in the horizontal
component oi SV or P. For certain angles of incidence,
the SV component arriving at the surface of the earth is
totally reflected. In this case the ground moves theoret-

168

Eartiiqtakes IX Kerx Couxty, 1952

Table 1. Aniplituile of S-uares and angle of polarisation (t).

fBull. 171

X, K aro nortli-south and east-west components of the ground motion in S ; rt and h are calculated ground amplitudes of SV and S// respec-
tively, all rounded off and in microns except for values indicated by *, which are given in arbitrary units. A = epicentral distance in de-
grees, u = azimuth at the epicenter towards the station, y ^^ azimuth at the station towards the epicenter: both are counted from north
towards east.

Station

Ottawa

Palisades- . -
Honolulu - . .

Weston

San Juan

Reykjavik.-

Kiruna

Sapporo

Aberdeen

Bergen

Sendai

Tokyo CMO

Kew

Uppsala

Matsushiro- .

De Bill

Copenliagen.

Hamburg

Coimbra

Eva Peron-.
Cliristchurcli

A degr.

I degr.

50
67

202
6.i
Slfi
28
10

.3i:i

310

307

310

.31

135

223

T degr.

265

275

270

—BO

295

—30

—48

—.53

—43

—36

—47

—41

—52

—45

H-5
— 10

— 3H

— 5

—4
—10
—25
-1-10
—30
—40

— 8H
—20
— 10

—3

+ 9

-1-25
—20*
+ 12
-(-30*
4-12.1^

-1-4

-F4

+ 7

-1-4
0±
0±
+ 17

-7}i
—30
+44

+ 4
+ 14
+ 15

+2
—15

— 14

+ 13*
+ l'A

— 18*
— 5
+i'A

— IWi
-i'A
— 4Ji
-3H
+ 1H
+3(4

— 16H

+ 7
—24
—33

— 5K
-13!^

— 10
—2
—4!^

— 13

+ 11*

+ 10
— 2H*
+ 5H
+ 3Ji
—3H
+ H
+ 1
-M

-IH

—4
— 5

0
+3

0
-Hi

—H

+ H
+ iH

—8

s degr.

—98

132

—12

— II

—53

—52

—8

—4

—43

—120

ically in ellipses. This theoretieally disagreeable condi-
tion occurs in shallow shocks rouwhlv at distances of
between 30 and 3000 km (about 20 and 2000 miles) from
the epicenter. For more details, see Gutenberg' (1952).
S-waves recorded at short distances from sources in
southern California have been used by Dehlinger (1952)
to study the g-round motion in a number of earthquakes.

In the main earthquake of July 21 the motion at short
distances was too large on the records to find iS. Conse-
(jueutlj', our investigation of recorded S-waves is limited
to epicentral distances greater than about 35°. Another
distance range which has to be excluded is between about
82° and 88° where <S7iiS' follows iS' immediately or pre-
cedes it by less than 25 seconds (figure 4 in Part II-6)
and affects the amplitudes of S too much for practical
use. Unfortunately, many European .stations with excel-
lent records are in this range of distances. Even bej'ond
88°, 8 is frequently atfected by SKS, SKKS and re-
lated phases. Finally, at distances beyond about 110^^ .'>?
gradually fades out. In the remaining range of distances,
all available records of stations with two horizontal com-
ponents having instrumental constants not to different
from each other and known relationship between direc-
tion of ground motion and direction of recorded waves
were carefully studied. Theoretically, vertical compo-
nents can be \ised, too, but instances of well recorded S
waves on vertical records are rare on account of the
usually small periods for which most vertical instru-
ments have their maximum magnification, and the fre-
quently small amiilitudes of the vertical component of S.

In table 1, amplitudes of the two horizontal com-
ponents N. and E. of S (positive towards north and
east respectively) are entered for those stations for which
the records fulfill the conditions mentioned above. If
the magnification factors of the seismograph are not
known, aiiii)litudes are given in arbitrary units and are
marked by "*". Otherwise, they are roughly ground
motions in microns in the first clear half S-wave. The
theory (Gutenberg, 1952) gives for the amplitudes a of
SV and b of SH in the incident S-wave

fl = (.V cos y -i- E sin y) /« = Z/w

b = (Ecosy — X sin y)/2 (1)

where y = azimuth at the station towards the epicenter
counted from north towards east ; Z = vertical com-
ponent (usually not used here), u and w are constants

Tahle 2. Approximate values used for quuittHies in equation (1).

Distance

100

degrees

18 degrees

1.7

1.8

1.9

1.0

0.9

0.8

0.7

depending on the angle of incidence i for a given ratio
of the wave velocities in P and .S' at the earth's sur-
face. Approximate values of these quantities as func-
tion of the distance A in degrees are given in table 2 on
the basis of averages found previously (Gutenberg 1952),
The polarization angle e (called "Sehwingungs-
winkel" by Galitzin, 1911) is the angle between .S' and
its component SV. From this definition it follows (fig. 3)
that

tan e = RH/8V = b/a

(2)

For earlier investigations see Galitzin (1911), Guten-
berg (1952) and Ingram (1953). If it is not certain
whether the first or a later half w'ave of S has been
measured, it may be preferable to count £ from -|-90°
to — 90°, otherwise it may be counted from -|-180° to
to — 180°. However, the same procedure should be used
for observations and calculations. If SV = 0, a ^ 0
and E = ±90; if SH = 0. b = 0 and e = 0 or ±180°.
Since e does not depend on the individual values of
a and b. but only on their ratio, absolute values of the
ground motion are not required for its calculation and
the data marked with asterisks in table 1 are just as
useful as the others provided that the instrumental con-
stants do not differ too much. The values of a and b in

Part II]

Seismolooy

160

150-

rCHRISTCHUfiCH

EUROPE AND

SAN JUA

EVA peron/

f rHONOLllLU

EASTERN

ry -

y sv-o-

120'

JAPAN
1 1

NORTH AMERICA

SO-

1 1

SO'

30*

, \

/*'

SH-O-

-30*

y^ .^^

60«

good • _
± o

■90'

/ •

5V0-

nn»

n_L. 1

a =A

zimuth oT epicenter -
1 1 1 1

1 1

270'

360'

Figure 4. Observed angles of polarisiUion e as a function of
the azimuth towards the station at the ejiioeuter in the main shock
of July 21, 1952. Curve 1 is calculated on the assumption of iliii-
slip along the fault, curve 2 on the assunipfion that the slip was
towards the north in the ui)per hlock. towards the soulli in llu-
lower hlock. separated h.v the White Wolf fault.

table 1 are supposed to give the •rround motion of the
first half iST and 8Ii wave respectively in microns, ex-
cept for the values indicated by asterisks, a is positive
if the horizontal component of 8Y is towards the epi-
center and its vertical component upward; & is positive
if its direction is to the right of the ray, looking from
above, e is calcidated from equations (2) and (1) or
directly from the ratio r ^ E 'N of the ground ampli-
tudes of the east-west and north-south components of S :

. u r cot Y — 1 ,„^

tanE= ^ — -^— (3)

2 r -(- cot Y

The azimuth y of the ray at the station and its azimuth
a at the epicenter are measured on a globe; u is taken
from table 2.

It is usually assumed (Gutenberg 1952; Ingram 1953)
that the polarization angle e does not change during the
propagation of an S-wave, or that the change in ampli-
tudes of SV and SH is the same, percentagewise. It is
then possible to plot e as a function of the azimuth a
at the epicenter for stations at about the same distance
in various azimuths. However, since the accuracy of the
observations is not very high, and since the distance
enters only through the angle of incidence i, which
changes relatively little in the range of distances in-
volved (see table 2), all results for e are plotted together
in figure 4 as function of a regardless of the value of i.

Apparently thus far no equations have been developed
to calculate e if the necessary quantities concerning the
fault and the direction of slip are given. Such equations
can be found by use of trigonometry. A more elegant
method is the following, given by Mr. John M. Nord-
quist :

The amplitudes of (1) P. (2) SH, and (3) SV as
they leave the source along a ray are assumed to be
proportional to the components of the displacement of
the fault in the direction of (1) the ray, (2) a horizontal
line perpendicular to the ray, and (3) a line perpendicu-

lar to the ray and to (2) respectively. The direction of
the ray is specified by the azimuth a of its vertical plane
and the angle of incidence i (measured from the verti-
cal). Let <l> and t|' be the corresponding angles for the
fault displacement.

In vector notation, the component of the vector Cc
in the direction of tlie unit vector d is the .scalar product

C (c-d) = C (o d., + c„ d, + c- rf,)

(4)

where the subscrijits t, y, z indicate components of the
unit vectors c and d in a rectangular coordinate system
(.r, y. z). Choosing coordinate axes ]>ointing to the north,
east, and down, we obtain the following schedule of
components :

Unit vectors

in direction

Fault motion

ray

Ampli-

Angle of

tude

factor

azmi.

incid.

-i

a-ttO''

90°

i + 90°

Components
North East Down

sin 'ii cos ^
sin )' cos a

sm a
cos J cos X

sin 4' sin <^
sin I sin a

- cos a
cos ) sin a

cos Ijl
cos i

- sin i

Substitution in equation (4) and collection of similar
terms lead to

p = A [sin \|' sin i cos (a — <^) -|- cos \\> cos i] (5)
6 = 4 sin \\> sin (a — <j>) (6)

a = A [sin \\< cos i cos (a — <^) — cos i|i sin i] (7)

All these equations of I\rr. Nordquist are based on the
assumption that the amplitudes of P and 8 generated at
the source depend only on the amplitude of the displace-
ment A on one side of the fault. Actually, the opposite
motion of the other block enters also. However, if the
ray starts in a direction which forms not too small an
angle with the fault surface at the source, the efl'eet of
the other side \ipon the first motion in S or P can be
neglected in a first approximation, that is, equations (5)
to (7) can be used if the station is far enough from the
fault plane circle in figure 2. This requirement is ful-
filled for all stations in table 1 with the exception of San
Juan, as may be seen from figure 2. For this reason the
point indicating the value of e for San Juan in figure 4
has been marked by ?. It also has to be considered that
other factors, for example the period of the generated
wave, affect the amplitudes of P and -S. However, since
SV and SH are components of the same S-wave, it may
be assumed that SV:SH = a:h.

Mr. Nordquist 's equations lead to

. , , cot tj» sin i ,-.

cot E = cos ( cot (a — d,) — — — -i (8)

^ ^' sm (a — <^) ^ '

For its application the various angles must be known.
For the angle of incidence of the rays an average of
i = 22° is taken (see table 2). The only other quantities
which have to be assumed for the calculation of a and 6
from equations (6) and (7) are the dip and strike of the
direction of motion. It was assumed first that the motion
was a pure dip-slip, and that the angles <^ and t[i are
given by the investigation of the compressions and dila-
tations. Consequently, it was assumed that in the lower
block (in which all raj-s to the stations in table 1 start)
•^ = 30° (more accurately, it should be 90 — 63 = 27°)

170

Earthquakes in Kern County, 1952

[Bull. 171

and <^ = 90 + 50 = 140° considering that the azimuth
of the fault trace is about 50°. The curve 1 is indicated
in figure 4. It is definitely too high.

There are various ways to calculate first approxima-
tions of the (luantities involved. From equations (6) and
(7) quadratic equations for cos (« — <^) or sin (a — <^)
can be derived which can be used to calculate (f) from the
data of each station. The equation for x = sin (a — 4>)
is of the form ax^ -{- bx -\- c =: Q, where
a = sin'-' \\i {y- -\- eos^ i)
b =z y sin - iji sin i
c = sill' i — sin- \\),
ii y = eot e. For calculations, it was assumed that
ij) — 34° and i = 22°. £ was taken from table 1 for each
station. The resulting values of a — <^ together with the
values of a to the respective stations give <^ := 11 ± 24°.

Some special values of e can be used to find approxi-
mations for <j> and \\' and, in addition, give information
on the way in which (f>. i(> and i affect the curve showing
e as a function of a. If & = 0, £ = 0 and 4> = a,,. Figure
4 shows that £ = 0, if a is slightly over 0°. This £ indi-
cates that <f> is slightly greater than zero, thus confirming
the calculations. For a = 0, e = 90° and equation (7)
gives

eot Tj» r= eot i cos (ago — a,))

Figure 4 shows that og,, — a,, is about 120°. This gives
\[) = 39° ±. However, in our case the cosine changes
rapidly with changing argument, so that this equation
gives rather rough results for i|i; it may be of greater
value in other instances.

From equation (8) we find by differentiation

dt
da

= sin - E

[cos i — cot op sin i cos (a — <j))
I sin Ma — <^)

(8a)

dt cos - E [ cos i — cot il) sin t cos (a — <^ ) ] , ov, \

Or-r- = } ■■ } -T T-, -. Tl (.^°/

rfa [cos t cos (a — 4>) — cot \p sin t[ ''

For E = 0 and (a — <^) =0, equation (8b) gives

cot Ui = cot t — —. r-r^ — , , . (8c)

Sin t (at / da)o

This may be useful for a rough determination of i^'.
Figure 4 shows that for e = 0, approximately {dt / da)o

= 0.7, which gives cot tj) = 1^ + , \J> = 34° ±. For e =
90°, equation (8a) leads to (dt / da)go = cos i.

On the other hand, for a given azimuth a of the strike
of the fault, the angle \|' depends on the angle (j>. If the
dip of the fault (counted from the horizontal plane, as
usual) is indicated by 8, it is found that

tan \|' = (cot 8) /sin {<(>

(9)

The angle |3 between the direction of actual motion in
the fault plane and that for dip-slip is found from

sin P = cos (<f> — o) sin aj» (10)

Finallv, curve 2 in figure 4 was calculated for <)> = 180°
(or zero) and <^ — a = 180—50=130°. With 8 = 63°,
eciuation (9) then gives \)i := 34°, as found above from
(rfE/rfa)o. Curve 2 is then given by

cot £ = 0.93 cot (a — <^) — 0.55/sin (a — <^) (11)

The resulting curve 2 in figure 4 fits the observations
beyond expectation considering the theoretical assump-
tions and the errors involved in the calculation of the
ratio of the EW — to the NS components of the ground
motion in S at the various stations. Before other investi-
gations of this type produce similar good agreement as
that in the present investigation, there is the danger of
over-confidence in the results of this new method.

Equation (10) gives |3 = 21°. Thus the observed com-
pressions and dilatations as well as the observed dis-
placements in the transverse waves combined with the
assumption that the strike of the White Wolf fault is
in an azimuth of about 50° from north towards east
lead to the following results : the fault plane has a dip
of between about 60° and 66° towards southeast
(S 40° E) at the depth of the source (about 10 miles) ;
the slip along the fault was roughly up towards north
in the upper block, down towards south in the lower;
the angle between the direction of slip and the direction
of dip is about 20° ; thus the motion was much closer
to dip-slip than to strike-slip; the vertical component of
the slip was about 1.4 times that of the horizontal; the
horizontal component produced a relative movement
northeastward in the upper block (southeast of the
fault), southwestward in the lower (northwest of the
fault).

8. MAGNITUDE DETERMINATION FOR LARGER KERN COUNTY SHOCKS, 1952; EFFECTS
OF STATION AZIMUTH AND CALCULATION METHODS

BV B. (JUTENEKRG

ABSTRACT

Mptho<l.s for magnitude tleterminatioii are summarized. Values
for the magnitude of the main shook are listed on the hnsis of
wave amplitudes measured on seismograms of individual stations.
About 2(H) data from body wave amplitudes result in a magnitude
of 7.H with only slight variation in azimuth. However, amplitudes
of surface waves at a given distance show a clear variation with
the azimuth in which they start with a maximum towards north-
east I in the direction of the fault) about 10 times the minimum
which is found in waves starting towards southwest. This is con-
sidered to be a consequence of the fact that in the main shock
the breaking proceeded northeastward from the neighborhood of the
southwest end of the active fault segment. In the largest after-
shocks there was no appreciable difference in the amjilitudes of
surface waves in those azimuths for which data are available. The
magnitude of the main shock determined from surface waves is
7.C to 7.7.

Magnitudes of the largest aftershocks are listed. They are cal-
culated from maximum amplitudes at near-by stations, from ampli-
tudes of body waves at distant stations, and from surface waves.
The differences between the various results for a given shock are
relatively small.

The mafrnitude of an earthquake was originally defined
by Richter (1935) for shallow shocks in southern Cali-
fornia as the logarithm of the maximum trace amplitude
expressed in thousandths of a millimeter with which the
standard short-period torsion seismometer (period 0.8
sec, magnification 2800, damping nearly critical) would
register that earthquake at an epieentral distance of 100
kilometers. In southern California shocks with focal
depths of about 16 km. magnitude 2 corresponds usually
to the smallest earthquakes which are felt ; average shal-
low shocks of magnitude 4^ to 5 may protluce small
damage, magnitude 5^ to 6 may cause an acceleration
of one tenth of gravity; shocks of magnitude 7 or more
are called ma.jor earthquakes, those of 7J or more, great
earthtjuakes. The largest magnitude found thus far for
earthquakes since 1904 (when sufficient instruments for
the determination became available) is 8.6. However,
there is indication that the Lisbon earthquake of 1755
may have had a magnitude of about 8^ or even slightly
more.

Gutenberg and Richter (1936) extended the magni-
tude scale to apply to shallow earthquakes occurring
elsewhere and recorded on other types of instruments.
The physical meaning of the scale was discussed, im-
provements were introduced and a nomogram for its
application (drafted by Mr. J. M. Xordquist ) was pre-
sented by Gutenberg and Richter (1942). If u is the
horizontal component of the ground amplitude of the
largest surface waves with periods of about 20 seconds
in shallow shocks of average focal depth (15 to 30 km),
then

M = P + S + log u (1)

where S is a small constant, different for each station,
to correct for local conditions, and F depends only on
the epieentral distance. Revised tabulations for ii^ as a
function of distance, and values of /S for a number of
stations were given by Gutenberg (1945a). Magnitudes
of shallow earthquakes were then correlated with ampli-
tudes and periods of waves through the earth recorded

at distances of over 1,000 miles (Gutenberg. 194.5b). Con-
sequently. P, PP and S are now also available for mag-
nitude determination. For a discussion of the relation-
ship between magnitude ,1/ and energy E of an
earthquake, sec Part 1 1-1.*

For the calculation of the magnitude M from records
written at distances of less than 1,000 miles the original
Richter method still has to be used : that is. trace maxima
on standard "Wood-Anderson seismographs have to be
measured and the magnitude then is found from a table
or the Nord(iuist nomogram. For distances greater than
about I.OOO miles either the maximum ground motion in
surface waves with periods of about 20 seconds is deter-
mined and a nomogram or tables for F in equation (1)
are used (Gutenberg, 1945a, p. 7), or values of ground
amplitudes a in microns of body waves having periods
T are inserted in the equation

M = log (a/T) +A + B + C. (2)

A in & function of the distance and may be taken from
Gutenberg (1945b, table 4, p. 65) for the various types
of waves or from corresponding graphs (Gutenberg,
1945c.). B is a station correction (usually not much dif-
ferent from S in equation 1 ; see e.g., Gutenberg, 1945c,
table 1). C is an empirical correction to be applied for
shocks of magnitude over 7 ; it is about 0.2 for the main
shock of July 21 and zero for all others. Xo correction
corresponding to C is used if the calculation of M is
based on amplitudes of surface waves.

If horizontal ground amplitudes u (given in microns)
of surface waves over the greater arc (W2. across the
antipodal point of the epicenter; see figure 7 in Part II-
1.), with periods T of about 20 sec are known, the
magnitude M is given by equation (1) where F (for a
giveu distance A) is taken from table 1. These values

Table 1. Average values of F in equation (11 for W2-\caves. A =
distance in degrees orer the shorter arc from epicenter to station.

100

120

140

160

6.9

6.8

6.7

6.6

6.4

6.2

5.9

5.4

of F are based on earlier research of Gutenberg and
Richter (1936, p. 120) and are rough approximations
since with increasing length of the wave paths the effects
of variations in structure of the earth's crust accumulate.
Records of the main shock written at distances less
than 18° were too large for measurement of amplitudes
on torsion seismographs and, consequently, all magni-
tude determinations for the main shock are based on
ground amplitudes calculated from records at distances
over 20°. Results are' given in table 2. A sign "±"
indicates that the constants of the instruments were not
well known, or that the trace amplitude was in doubt, or

• In a letter to \ature (1955) Gutenberg and Richter have re-
discussed the definition of magrnitude. For the main shock the
new magnitude M is 7.4 from body waves as well as from
surface waves. Revised equations and calculations give a cor-
responding energy release of 4 X lO^" ergs. The correction C
in equation (2) is no longer added, but a correction has to be
used if M is calculated from surface waves.

(171 )

172

Earthquakes in Kern County, 1952

[Bull. 171

Table 2.

Magnitudes determined for main shock. July 21, 1952. from direct longitudinal waves (P). longitudinal waves reflected once at the earth's
surface (PP), direct transverse waves (S). maxima of direct surface waves (Max) and maxima of surface waves over the greater arc (W2).
H = horizontal component. Z = vertical component. Corrections B and C in equation (2) are added. 4 = epicentral distance in degrees. An
asterisli (•) indicates use of reported amplitudes.

Station

Magnitude determined from

PPZ

PPH

Max H

W2H

18.6

19.3

19.8

23.4

24.4

25.1

25.4

23.7

30.0

34.3

35.7

36.6

37.7

41.1

41.5

42 Q i

62 I

62 .-.

69.7

70.4

73.0

73.3

74.0

74.3

74.6

75.7

76.1

78.3

78.5

78.9

80.5

80. G

81.4

81.6

81.7

81.8

82.1

82.3

84.0

84.3

84.4

85.0

85.1

85.5

85.8

86.3

86.4

86.6

86.7

86.9

87.3

87.4

88.7

89.1

89.3

89.7

89.9

91.5

92.5

97.2

99.6

99.9
103.7
108.3
109.7
114.6
114.9
117.2
125.2
125.2
125.4
132.4
134.6
160.1

7.9

7.6
7.5*

7.7
7.6d=

7.8

7.2
7.6±

7.4±

7.8

7.9
7.3

7.3
7.2

7.5
7.7

8.0

7.4

7.3

7.5

7.0±
7.2±

7.1±

7.4
7.4

7.3
7.5
7.2

7.4±

8.1

7.3±

7.4

7.8

8±

7.6

7.4

7.6

7.6±

7.8

7.7

7.7
7.5

7.5

8.0±

7.5

7.3

7.4±

8.1

8.0

7.9

7.2

7.0

7.6

7.8

7.4

7.3

7.4

7.5

7.6

8.1

7.3

7.7

7.8

7.7*

7.4±

7.5

7.6

7.8d=

8.1±

8.0

8.1

7.9*
7.6
7.7*
7.0

7.4
7.7
7.7
7.6
7.6
7.3

7.4

7J^

7.4±

7.9
7.3

7.2
7.7

7±

7.6
7.4

7.3

7.4

7.1

7.2
7.1

7.2
7.4

7.4±
7.5

7.3

7.4

7.4
7.6

7.2±

7.1

7.6
7.4

7.0
7.0
7.3
7.5

7.6

7.3

7.2

7.6

7.6
7.8
7.8
7.3

7.3

7.2

7.6

7.3

8.0±

7.3

7.5

7.3±

8.0±

7.3±

8
7.5

7.4

7.4
7.8

7.3±
7.3

7.3

7.6
7.3

7.4

7.6

7.4±

7.6

7.3

7.6d=

7.3
7.9
7.9
7.8
7.1

7.8
7.4
8.1
7.9
7.9

7.3±

7.9

7.3±

7.6

7.3

7.6

7.4±

7.1

7.8±

7.1±

7.9±

7.6

7.9

7.8±

7.3

7.7*

8.1?

7.6

7.5±

7.3±

7.3

7.2±

7.8

7.2±

7H±

7.4±
7.5

7.8
7.7

7.7d=

7.9*

7.2

7.3

7.4

7.6

7.9

7.9
7.8
8.0
7.8

7.8

7.9
7.8±
7.8
7.9

7.9
8.0±

7.6±
7.9±
7.9

7.9*

t
7.7*
7.9±
7.9*

7.9
7.6
7.9

7.9

7.4

8.0

7.1

6.8

7.0

7.7

6.9±

7.3

7.9±

7.6

7.5

7.6±

7.4

7.2

7.8±

Puebla - - - -

rhicano USCGP - --

Ottawa - -

8±

Halifax .- -- - -

7.6

La Paz - -

7.7

6.9±

6.6±

6H±

Tokyo ERI --. - - -

7.5±

Tokyo CMO - - -.

Kew -- -.- ---

7.1±

6.8

De Bilt -

7>i±

Osaka -- --

6.8±

6.7±

7.0±

7.1 ±

Cheb -. - --. --. --

6.7±

Tortosa -

Chur - - - -

Rtara Dala

6.9

Skalnate Pleao

Trieste --

6.9

Prato -

Eva Peron (La Plata)

7.4±

Zagreb

7.4±
7.2±

Beograd

7.2±
8.8

8.2

8.3

8.4

7H±

8.0dt

Quetta -

7.3±

7.8±

7.3±

7.1

7.8

Tananarive- ,. -

Part Til

Seismology

173

(especially for 5! near 84^ and for ]V2) that there was
doubt about the proper ideiititieation of the phase. The
resulting values of M depend to some extent on local
conditions at the source and the station (unless cor-
rected for), and on the wave path. For PP, the longi-
tudinal wave reflected at the earth's surface (or at the
Mohorovicic discontinuity) about half way between
source and station, conditions at the point of reflection
also affect the recorded amplitude ; there is good indica-
tion that reflections under the bottom of the Pacific
Basin result in smaller energy for PP and probably
correspondingly greater energy in PS than those under
continents. No corrections for these effects are applied
in table 2 and. consequently, magnitudes calculated from
PP may be expected to be too small in instances of re-
flections under the Pacific Basin.

Average magnitudes calculated from amplitudes of
the vertical (Z) and horizontal (H) components of the
phases P, PP and 8 are listed in table 3. The differences
between the results given in the five columns are rather
small. In finding the most likely value of M we have to
consider that the magnitude resulting from PP is prob-
ably slightly too small since no correction was applied
for reflections in the Pacific Basin that usually lead to
relatively small amplitudes of PP. In addition, the mag-
nification of many old vertical instruments is likely to
be smaller than given by the stations (C4utenberg, 1945c,
p. 119), and consequently slightly too small calculated
values of M are to be expected if PZ and PPZ are used.
Thus we may conclude that the magnitude indicated by
the body waves of the main shock on July 21 is near 7.6.

Table 3. Magnitude .1/ of main shock on July ^1. calculated from
amplitudes of hody icares. n ^= number of stations.

HELWAN

A =109 7°,

Phase and component

PPZ

PPH

7.5
29

7.6
62

7.4
16

7.5
34

7.6
53

The average magnitude calculated from P on records
of stations in the azimuth towards Japan is 7.8, whereas
the corresponding value of .¥ found from records at
stations in the direction towards northern Europe is
onh' 7.4 to 7.5. Magnitudes calculated from PP show a
distribution in azimuth similar to that in P -. 7.6 in the
azimuth towards Japan and 7.4 towards Europe. Prob-
ably these differences are mainly a consequence of effects
already mentioned, but, in addition, may be expected to
include a term produced by the distribution of energy
with azimuth at the source which depends on the direc-
tion of strike and dip of the fault surface at the depth
of focus, on the direction in which the faulting pro-
ceeded and on the direction of motion along the fault
surface, and is different for longitudinal and transverse
waves.

Magnitudes calculated from SH (table 2) show a rela-
tively high average of 7.7 in the direction towards south-
ern Europe, 7.6 towards northern Europe and Japan,
and only 7.4 towards South America.

While the azimuthal differences in the magnitude cal-
culated from P, PP and S, though fairly consistent, are

c:iy:i^S5^^>!22^3jSSii=::>c

-^ssrszs^E^^^.

M, 12:45
WELLINGTON N - S, A = 97.2°, a = 224°

•.v--v-.''*v ■/,,• >-"-■

M, 12:48

...^ I —

Wg , 14:08

FlGlRE 1. Parts of Milne-Shaw records of main shock July 21,
showing nia.xima of direct surface waves .1/ and of IFj. The length
of 1 minute is indicated in each record. All instruments have the
same free period (12 sec.) and the same magnification (250).

rather small and may be produced by accidental ac-
cumulation of errors, the surface wave maxima show a
very strong dependence on the azimuth of the station.
Whereas maxima recorded in Europe lead to a magni-
tude of 7.9 ± 0.1, those recorded in New Zealand and
Australia give only about 7.1. About fifty times as mtich
energy was radiated in the surface waves towards north-
east than towards southwest. This is confirmed by the
fact that on all records of the main shock written in New
Zealand and Australia the largest surface waves over
the greater arc (ir2; see Fig. 7, Part II-l) have ampli-
tudes as great or greater than the largest direct surface
waves (examples in figure 1) although the paths of the
ir2-waves to these stations are about twice to three times
as long as those of the direct surface waves. On the other
hand, ir2-waves cannot be identified on most European
records, since in Europe their amplitudes are not sig-
nificantly larger and possibly even smaller than those
of late direct surface waves arriving simultaneously
with ir2, \\ to 2 hours after the maximum of the direct
surface waves (depending on the epieentral distance of
the station). Most magnitudes calculated from 172 for
European stations are marked "±" in table 2 since
there was doubt whether the corresponding waves were
actually ir2-waves or late direct waves.

Figure 2 shows calculated magnitudes M, and ampli-
tudes 048 of surface wave maxima which would have
been observed at a distance of 84° (at 84°, log a = .¥ —
5.0) as a function of the azimuth towards the stations. No
correction was made for the effect of wave paths (see
e.g. Gutenberg, 194oa, p. 9) ; these would increase the
values of iV and a for Japanese and South American
stations (azimuths near 310° and 130° respectively).

From fig. 2a average values of a were calculated for
nine different azimuths, and the method of least squares
was applied to calculate a sinusoidal curve fitting the
data best. The resulting curve for the amplitude as4 in
microns (including standard errors) is given as function
of the azimuth a by

as4= (417.8 ±20.3) +

(342.2 ± 23.7) sin (a — 54.8° == 5.8°). (3)

174

N_^^

Earthquakes in Keen County, 1952
W N

[Bull. 171

600
FROM
Mai

W2 \»^

•

T GOOD ^v

V ± \

FROM REPORT

0°

90'

TT ?rTT2?0°

Figure 2. a, Calculated magnitudes of the main shock, and 6,
calculated amplitudes at 84" epicentral distance; hoth are hased
on maxima of surface waves (left scale) and of Wi waves (scale
at right) and plotted as function of azimuth k at the epicenter
towards the staticm from north (0") towards east (ilO").

The magnitude corresponding to the average amplitude
of 417.8 microns at a distance of 84° is 7.62 ± 0.02, the
magnitude corresponding to the average energy is 7.68
± 0.03. Thus, the recorded amplitudes of the surface
waves indicate that the magnitude of the main earth-
quake was between 7.6 and 7.7.

P S '

,;, ^

•ONE HINUTE — I

PASADENA STRONG MOTION, V = 4, 1952 JULY 21

Figure 3. Portion of record of .strong-motion seismographs at
Pasadena on July 21, 1952. Epicentral distance is 124 km. Record-
ing becomes visible when trigger device intensifies light. Small after-
shocks produced short almost straight lines about 15 minutes after
main shock (line below main seismogram) and HO minutes after
main shock (second line below shock). Length of 1 minute on
original seismograms is 60 mm.

Table i. Selected values of the magnitude M, calculated from
surface wave inajima for the aftershock of July 29, 7".

Direction from epicenter

Northeast

Southeast

Southwest

Northwest

Copenhagen

6.3

La Paz 6.1

Christchurch 5.9

Matsushiro 5 . 8

De Bilt

6.2

Huancayo 5.6

Wellington 6.1

Bombay 6 . 2

Kew

5.9

Puebia 6.2

Riverview 6.0

Sitka 6.0

Durham

6.1

Veracruz 6.2

Uppsala

6.0

San Juan 5.7

Praha

6.1

Aberdeen

6.1

StuttEart

6.3

Kinma

6.2

Roma

6.2

Hamburg

6.4

The trace amplitudes recorded by the WE and SN
components of the strong motion instrument at Pasa-
dena (fig. 3) are 70 and 50 mm respectively. Using
calibration data found by Gutenberg and Richter (1942,
p. 167) this gives a magnitude of 7.5± (log average
trace amplitude mm -\- log b = 1.8 -|- 5.7, corresponding
to A := 124 km). The corresponding maximum horizontal
ground motion at the Seismological Ijaborator.y is of the
order of 20,000 microns with a period of about 10 sec-
onds. These waves are superposed by smaller motion
with shorter periods.

The azimuth 55 ± 6 degrees in which according to
equation (3) the largest surface waves were recorded,
agrees with the .strike of the White Wolf fault (a = 50°)
within the limits of error ; the location of the aftershocks
relative to the main shock (fig. 1, Part II-6A) indi-
cates that the breaking was propagated roughly towards
northeast. The fact that in this direction the surface
wave maxima were roughly ten times the maxima in the
opposite direction (azimuth 230° zh) may be considered
to indicate that the speed of propagation of the frac-
turing was not much less than the wave velocity, so
that it was possible for each wave to increase along the
line of breaking whicli probably measured several tens
of wave lengths. However, it is difficult to develop a
detailed quantitative theory, since surface waves are
formed by a variety of complicated processes. A qualita-
tive discussion of these phenomena is given by H. Benioff
in his section. Mechanism and Strain Characteristics.

The more or less horizontal direction towards north-
east in which the faulting process proceeded should have
influenced the amplitudes of P and iS' as function of
azimuth much less than those of the maxiina at distant
stations, since the rays of the body waves arriving at
distant stations form angles of 50° or more with hori-
zontal direction.

In the aftershocks the faulting can be assumed to
have proceeded along much shorter distances than in
the main shock, and consequently no appreciably greater
amplitudes of surface waves in the direction of faulting
can be expected than in the opposite direction. Unfor-
tunately, records for the surface waves at distant sta-
tions are available in a variety of azimiiths for the shock
of July 2i), 7'' only. Characteristic values of M calcu-
lated from surface wave maxima for this shock are given
in table 4 ; the average of all calculations is 6.1 (table 5).
The calculated magnitudes show no difference in azimuth

Part III

Seismology

175

Table 5.

Magnitudes M of largest aftershocks in Kern County, 1952. (a>
averages calculated from trace amplitudes recorded by standard tor-
sion seismographs at near-by stations; (b) calculated from ampli-
tude of body waves, and tc) from annplitudes of surface wave max-
ima at distant stations; (d) averages from all data for the shock,
n =: number of data.

Date

hourimin.

July 21

12:02

5.6

12:05

6.2

6.6

6.4

12:19

5.3

19:41

5.6

5.5